[Safe Board] - [YP Home] - [Project Status] - [YP Wiki] - [Burichan] [Futaba] [Gurochan] [Photon] - [Home] [Manage]

[Return]
Posting mode: Reply
Leave these fields empty (spam trap):
Name
Link
Subject
Comment
File
What is the theme of this website?
Password (for post and file deletion)
  • [General Discussion] [Homepage Comments]
  • Supported file types are: GIF, JPG, PNG, PSD, TXT
  • Maximum file size allowed is 5000 KB.
  • Images greater than 200x200 pixels will be thumbnailed.

File: 2014-08-23 11_52_33-A Kiss For The Pytals.png -(896385 B, 808x627) Thumbnail displayed, click image for full size.
896385 No.4064  

Over the last week, I received support requests from individuals who want to translate the first visual novel into French and Polish. Both languages use accented Latin characters that the text engine does not support, and although they are not strictly necessary for clarity, it's one of many limitations of MScenario that needlessly complicates translation efforts. To that end, I have begun work creating a framework for Fuguriya's visual novels in Ren'Py that will contain translated text in plaintext files; the user will need to provide all other assets themselves. This framework will expedite translation work into all languages supported by Unicode and eliminate all of the shortcomings in the visual novel engine still present in the latest release, My Sworn Love For You.

To this end, I require a method of extracting all image, sound, and positional data for each line. ArcTool only extracts text and not associated graphic and voice files. While I can examine decrypted MSD files for this information in a hex editor, manually inputting them into Ren'Py commands will be prohibitively time-consuming. If anyone here can create a solution to rapidly parse MSD files for this information, including scene changes, filename references, and positions, and output all information into a file that I can easily convert into Ren'Py commands, I would greatly appreciate the assistance and begin work on doing this for all of Fuguriya's visual novels.

>> No.4067  
File: 2014-08-24 00_46_46-A Kiss For The Pytals.png -(913151 B, 808x627) Thumbnail displayed, click image for full size.
913151

For anyone interested in automating the process of extracting program flow data from MSD, my current workflow is as follows:

  1. Use ArcTool to extract the English text from the current translation patch, then insert them into a new RPY file. Dialogue brackets 「」 are stripped as they are not needed in Ren'Py.
  2. Open a decrypted MSD file in a hex editor.
  3. Check for background image changes. (e.g. bg03a)
  4. Check for voice and sound effect file references and insert them before the corresponding lines. I currently do this by manually comparing the dialogue in the decrypted MSD against the RPY file. (e.g. NANA0001, SE001)
  5. Check for image file references and insert them before the corresponding lines. (e.g. tna01s)
  6. Insert respective Ren'Py commands around each reference. With regular expressions, I can create the proper commands for each type of file quickly. (e.g. NANA0001 is replaced with voice "NANA0001" and tna01s is replaced with show na tna01s)
  7. Pairs of image references are given transforms that I manually defined as xcenter 0.25 and 0.75. (e.g. show na tna01s at l \ show yu tyu01s at r)

At present, I have recreated S001 of the first visual novel with voice and images. I will focus solely on converting the scripts to Ren'Py at this time, after which I will build a screen to roughly imitate Yurin Yurin's visual novels, not the original visual novels.

>> No.4068  
File: strings_conv_002.txt -(18991 B, 0x0) Thumbnail displayed, click image for full size.

I ran it through GNU strings and got a whole lot out of it (I'm not sure if it's all though). I used some vim commands over the whole document to make it to spec. It looks like it just needs a few more adjustments.

>> No.4070  

If it's good like that I'll redo this and do the rest once you let me know what to do with the anomalies (like the heart symbol, newlines and the GS/DAMMY lines is what I've seen so far).

>> No.4076  
File: s01.txt -(28703 B, 0x0) Thumbnail displayed, click image for full size.

>>4070
Thank you very much for your assistance.

The DAMMY and @newline@ strings should be stripped. The GS_ strings indicate transitions, so I'd like to have them left in for easy reference after I implement them in Ren'Py. Full-screen images like bg03a and EV02 should be preceded by scene . The \ character is meant to signify a newline as these commands are separate. Music files are always (M[0-9]{2}) and should be formatted as with play music "common/\1.ogg". Likewise with sound files (SE[0-9]{3}): play sound "common/\1.ogg".

I have not yet moved ahead of this file yet, so I will likely reach another special character that will need attention. That said, the heart symbol in hex appears to be 00 01 64 00 00 00 03.

Attached is my work-in-progress script for s001. Character definitions can be inserted with a regular expression determining the voice file prefix immediately preceding the line. The character defines are as follows:

n = NANA
y = YUNA
ga = GALA
gb = GALB
gc = GALC
tt = KYOS

There are also names for "子供A" and "子供B", but they have no voice files. I have defined them as ca and cb respectively.

>> No.4077  
File: strings001.txt -(28499 B, 0x0) Thumbnail displayed, click image for full size.

>>4076
Here is my S001. I'm not sure I can tell when a character should have code "q" or "ca" and "cb". I saw a few missing voices near the start of your version.
Is CU01 really "show cu01"?

>> No.4078  
File: 2014-08-25 08_42_40-A Kiss For The Pytals.png -(438382 B, 808x627) Thumbnail displayed, click image for full size.
438382

>>4077
Character "q" is "???" before a character's name is introduced, so it has to be manually defined or determined from hex code. Since it's seldom used, the former is viable. Characters "ca" and "cb" do not appear until much later.

I use show cu01 and not scene cu01 as scene clears all images, and the cu images are transparent overlays, not backgrounds.

What commands are you using to generate this output?

>> No.4079  

I got all the special characters (they were SJIS): ♡ ♪ 〜 “ ” … and おしまい at the end.
Here is what I have so far:
http://www.mediafire.com/download/d0ddto5bnncz5cl/strings.rar

>> No.4080  

>>4078

>or determined from hex code

I didn't see it yet.

>I use show cu01 and not scene cu01

All right. And capitalization?

This bash script, no manual editing:
#! /bin/bash
for msd in S0??.MSD; do
cat "$msd" | LANG=C sed '
s/\x81\x99/HARTEMOTe/g
s/\x81\xf4/NOTEEMOTe/g
s/\x81\x60/LONGTILDe/g
s/\x81\x75/QUOTEOPEn/g
s/\x81\x63/ELIPSEs/g
s/\x81\x76/QUOTECLOSe/g
s/\x81\x40\x81\x40/NEWLINe/g
s/\x81\x40\x82\xa8\x82\xb5.*/OSHIMAi/' | strings -eS -n 3 | \
sed '
s/NEWLINe/ /g
s/ */ /g
s/"/\\&/g;s/^ //
s/^[\\A-Z.].[a-z.].*/"&"/
s/^[^"].\.\.\../"&"/
s/HARTEMOTe/♡/g
s/NOTEEMOTe/♪/g
s/LONGTILDe/〜/g
s/QUOTEOPEn/ “/g
s/QUOTECLOSe/” /g
s/ELIPSEs/…/g
s/OSHIMAi/The End/g
s/^CU[0-9]*/show &/
s/^bg.*/scene &/
s/^EV[0-9]\+.*/scene &/
s/^BLACK$/scene &/
s/^SE[0-9]\+.*/play sound "common\/&.ogg"/
s/^M[0-9]\+.*/play music "common\/&.ogg"/
s/^[A-Z]\{2,\}[0-9]\+$/voice "&"/
s/^t\([a-z][a-z]\).*/show \1 &/
s/ -\([a-zA-Z]\+\)-\([. ]\)/ {i}\1{\/i}\2/g' | grep -v $'\xff\xff\nDAMMY' | \
sed '
/^show/{N;s/\(show.\)\n\(show.\)/\1 at l\n\2 at r/}
/^voice/{N;s/\(voice "\(....\).\)\n\(".\)/\1\n\2 \3\n/

s/YUNA "/y "/
s/NANA "/n "/
s/GALA "/ga "/
s/GALB "/gb "/
s/KYOS "/tt "/

}' | \
sed '/^"/{:more;N;s/\n[^"]/&/;Tmore;s/\n[^"]/\n&/}' | \
sed '
s/^/ /
s/- \(chaa*n\)/-\1/g
s/- \(samaa*\)/-\1/g
s/- \(senpaa*\)/-\1/g' > "strings_$msd.txt"
done

>> No.4081  
File: script.txt -(1192 B, 0x0) Thumbnail displayed, click image for full size.

>>4080
I'll attach the script, that didn't work exactly.

>> No.4082  

>>4080
The heart symbol begins occurring in the fourth visual novel, Beloved Photograph, onward. The cu files are lowercase, as is the p in EV__p1 files.

The scene EV commands appear to be duplicated. If possible, strip duplicates of background file references. Strip also MSCENARIO FILE and the last three lines after the last line of dialogue. The eyecatches after every three chapters don't appear to be included in these files, so they will have to be manually implemented as well.

>> No.4083  

>>4082

>Character "q" is "???" before a character's name is introduced, so it has to be manually defined or determined from hex code.

Do you know what the hex code is?

>If possible, strip duplicates of background file references

I will do the EV's. I didn't see any for "scene bg".
At the end of S015 there are two in a row, but they're different:
scene EV20B
scene EV20

Can "GS_EN_NS" go?

Should I add "pause 1.0" after every "scene BLACK"?

>Strip also MSCENARIO FILE and the last three lines after the last line of dialogue.

The end of S015 with おしまい/"The End", too?
All the scripts should still end in "scene BLACK", right?

>The eyecatches after every three chapters don't appear to be included in these files, so they will have to be manually implemented as well.

Do you know what the hex code is?

>> No.4085  
File: EV20B.png -(2015716 B, 1600x1200) Thumbnail displayed, click image for full size.
2015716

>>4083
I do not know where to check for name references in hex. Eyecatches always appear after every third chapter, so we do not need to search for them. They can be implemented separately and jumped to after chapters 03, 06, 09, and 12 and before the following chapters.

EV20B (attached) is simply a larger version of EV20. I haven't checked the original behavior, but I believe it's used for scrolling.

Each play music command should be preceded with stop music fadeout 1.0. Add pause 1.0 after scene BLACK as well.

The GS_ files are masks for transitions; they are found in MGD. GS_EN_NS is a simple iris-in transition. GS_CLR is a horizontal-in transition used after every image change. I believe we can do away with it, as some of the newer visual novels do not use it at all. I'd like for the transitions to be commented out with # for the time being.

The branches at (sel0[1-9]p) require special attention as well. Please comment them and the following file sel02_c.bmp out for the time being, and add an additional comment #TODO: Branch.

These issues aside, I have been able to run these files in my project with minimal modification. Thank you again for your help.

>> No.4086  
File: script2.txt -(1610 B, 0x0) Thumbnail displayed, click image for full size.

>>4085

>These issues aside, I have been able to run these files in my project with minimal modification. Thank you again for your help.

That's great. You're welcome!

New version:
http://www.mediafire.com/download/31pjn891ahizibd/strings_v2.rar

Script (v2) attached.

>> No.4094  

>>4086
I'd like you to also process the MSD for the remake version as well. I need to see the differences in music and graphic cues.

>> No.4098  

>>4094
How do I arctool the remake? The preset for SH1 doesn't seem to work.

>> No.4099  

>>4098
For reasons unknown to me, newer builds of ArcTool label the remakes as "German Hanabira". Use this build listed on the wiki: http://www.mediafire.com/download/mlq46si847j77ja/ArcTool_11D.zip

I've also extracted the remake MSD already.
http://www.mediafire.com/download/m75wdkwp9u7rjdg/MSDr1.zip

>> No.4101  

>>4099
All right, it seems to be working with a small adjustment. Will this do?
http://www.mediafire.com/download/y5kgtqc71b4z7l5/Remake_strings.rar

>> No.4104  

>>4101
Yes; the Japanese output will also allow me to add Japanese language support.

Does your sed script change the dialogue brackets into quotation marks? I would like to preserve them in the Japanese script. Additionally, when a cu frame is cleared from the screen and character images are to be shown in its place, I will need it to hide before the next show statement for the characters, or they will remain on screen over the characters. Will it be possible to add this to your script?

>> No.4105  

>>4104
I noticed that the heart emote was actually a star emote ☆ in SJIS, so that should be changed in the English one as well, right?

>Does your sed script change the dialogue brackets into quotation marks?

Yeah, I'll change it back.

>I will need it to hide before the next show statement for the characters

For the English one, too?

New version, remake JP:
http://www.mediafire.com/download/gfnikxyfb96ab1a/Remake_strings_v2a.rar

>> No.4107  

>>4105

>I noticed that the heart emote was actually a star emote ☆ in SJIS, so that should be changed in the English one as well, right?

Yes. SJIS doesn't include a heart character, which is why the graphic is used instead, problematic as it is. ArcTool uses a different star character ★ to represent the heart graphic so as not to conflict with ☆.

The first matching string is the chapter title. Format it with $ save_name = "\1". I will do more testing to further tune the sed output to match Ren'Py code.

>> No.4108  

I'm comparing the two versions and there is a line missing in the remake:
S003
voice "YUNA0178"
y "You mean it? You'll call me Onee-sama?"

There's no trace of it even when I hexedit S003.MSD.

>>4107

>ArcTool uses a different star character ★ to represent the heart graphic so as not to conflict with ☆.

So I should make it ♡?

>> No.4109  

>>4108
It's right before this line:
voice "NANA0210"
n "「はい……」"

>> No.4111  

>>4108
Several lines have been changed in the remake, which I wrote about in 2012, after which you have reviewed. I do not know if the omission of this line was intentional.
http://axypb.net/changed-lines-in-a-kiss-for-the-petals/
http://nsfw.yuriproject.net/res/854.html

>So I should make it ♡?

Please do, but only for the visual novels that use it, which do not include the first three. Leave the ☆ intact.

>> No.4113  
File: 2014-08-28 11_27_05-A Kiss For The Pytals.png -(572021 B, 808x627) Thumbnail displayed, click image for full size.
572021

>>4086
Your current script omits a replace command for GALC. There are also some voice files that add an extra letter to the end of the file name, such as YUNA0267A. Please update the search expression for it: ([A-Z]{4}[0-9]{4}[A-Z]?)

The tilde character you used in your script 〜 appears to be incompatible with the default font Ren'Py uses. For compatibility, please change it to the ASCII tilde ~.

I have also noticed that Ren'Py does not clear previous images of a character when another is shown by default. In such instances where a single show command is followed later by a show command for another single character, the previous character needs to hide.

I am using individual RPY files for each chapter. At the start of each file, insert label sXX:, unindented, where XX is the number of the script, and at the end add indented jump sXY, where XY is the number of the next script.

ev12 is originally named in lowercase, and it's a tall image that scrolls bottom to top in the original visual novel. I will need to implement this separately. Please ensure that the lowercase is preserved, and add a comment #TODO: ev12 is a scrolling image. Later visual novels repeat this procedure.

Please run the newest version of your script on the English MSD again.

>> No.4114  

>>4111

>Several lines have been changed in the remake, which I wrote about in 2012, after which you have reviewed.

Oh, that's right. I think the other lines were just active/passive voice changes and not very relevant. Should I look over it again?

>I do not know if the omission of this line was intentional.

It's surely an error.

>>4113

>In such instances where a single show command is followed later by a show command for another single character, the previous character needs to hide.

Just 'hide'?

I haven't had the chance to review things closely the last few days, but I'll do it today or Sunday.

>> No.4115  
File: 2014-08-29 10_08_33-A Kiss For The Pytals.png -(939532 B, 808x627) Thumbnail displayed, click image for full size.
939532

>>4114
hide <character image>

There is the possibility that all characters exit the screen and are not replaced. I believe these are called with GS_CLR statements without matching characters. In this case, hide all characters.

If possible, I would like a complete retranslation of the script of the remake to be used retroactively. I have not been able to contact CountPacula for anything regarding the previous translation.

I created a rough approximation of the Yurin Yurin message window for demonstration purposes. The font used here is FreeSans, which includes support for most modern languages.

>> No.4116  

Hello, thanks for your work I appreciate that.

I have a question regarding dialogues. You wrote that dialogue brackets 「」 are stripped so should I remove them from my translation and not use them anymore?

>> No.4117  

>>4116
Yes, they will no longer be needed in the translations under this framework.

>> No.4120  

>>4116
Can you post Nanami's name and first line of speech in Polish?

>> No.4122  

Name and last name aren't changed in Polish if someone isn't talking about that person so it's the same as original, Oda Nanami.

And here is first line in Polish: Wydarzyło się to w kwietniu, gdy zakwitały wiśnie.

>> No.4123  
File: 2014-08-30 08_40_47-A Kiss For The Pytals.png -(943169 B, 808x627) Thumbnail displayed, click image for full size.
943169

>>4122
Further progress on the message window.

>> No.4124  
File: 2014-08-30 01_41_03-A Kiss For The Pytals.png -(919517 B, 808x627) Thumbnail displayed, click image for full size.
919517

>>4123
More samples: https://twitter.com/AXYPB/status/505697450242621440

>> No.4125  

I see in Polish all letters are fine.

>> No.4126  

>>4125
How much of the script have you translated so far? Ren'Py has a built-in method for providing translations with a set of files containing strings to be translated. They will need to be inputted manually. If you still have much of the script left to translate, you can resume in these Ren'Py translation files.
http://www.mediafire.com/download/n43nlf9qof7v0bq/SH1_polish.zip

>> No.4127  

I barely started, only 60 lines translated.

>> No.4128  

# game/s01.rpy:9
translate polish start_5c68a151:

# "It happened in April, as the sakura trees danced."
""

I'm guessing that I should put translation in empty "" beacause # is prabobly setting line as comment and won't be displayed?

>> No.4129  

Oh, and one more thing, can I also get menu files in Ren'Py for translation as well?

>> No.4130  

>>4128
Yes, insert your translation in the empty "". Remember not to include 「」.

>>4129
I am still in the process of building menus and screens, so I have not looked at the default structure files in detail. However, I have added the files containing default menu strings for you to translate in the meantime, screens.rpy and common.rpy. Note that I will add and remove from this list in the future.
http://www.mediafire.com/download/n43nlf9qof7v0bq/SH1_polish.zip

>> No.4134  

>>4115

>hide <character image>

You mean like "hide tna03s", right?

>There is the possibility that all characters exit the screen and are not replaced. I believe these are called with GS_CLR statements without matching characters. In this case, hide all characters.

I see, but they should then also come before the show lines, I suppose. Right now they are in between show and voice, which makes no sense. And there are a lot of them. They might be covering all the cases you want. (a 'hide character image' between character to character)

>> No.4135  

>>4134
The original script clears all character images on every image call using GS_CLR. I have found some cases where two characters are in a line, but the following line has only one character, and the exiting character isn't cleared. In those cases, a hide statement is needed for the image immediately before the next show statement. This may become further complicated in scenes in later visual novels with three characters on-screen.

>> No.4136  
File: 2014-09-01 02_44_32-A Kiss For The Pytals.png -(729138 B, 808x627) Thumbnail displayed, click image for full size.
729138

>>4135
An example of this occurs in s011, at line 486 of the corresponding text file you uploaded:

    show yu tyu01s at l
show na tna01s at r
#GS_CLR
voice "YUNA0597"
y "I'll go get us something to drink, if you don't mind waiting a bit."

voice "NANA0655"
n "Ah, alright."

show na tna01s
#GS_CLR
play sound "common/SE010.ogg"
stop music fadeout 1.0
play music "common/M02.ogg"
"I found myself left alone in Onee-sama's room."

After Nanami's line, Yuuna's image tyu01s should be hidden, but it isn't, so the image is still on screen and the positions of neither image change.

>> No.4137  

>>4136
I updated it:
http://www.mediafire.com/download/0x5caadiijr42mj/JP_SH1_Remake_strings_v3b.rar
http://www.mediafire.com/download/u0zz362f15a4g0q/EN_SH1_Original_strings_v3.rar

It's not doing anything with GS_CLR right now. It only hides right before a single character 'show', either when it's coming from a multi character 'show' or from a single character 'show' of a different character.

Where should the jump at the end of S015 go to? Currently I set it to S016.

>> No.4141  

>>4137
Set the end of the last file to jump ending. For the endings, I will require cleaned versions of every credits image, which I will use as backgrounds for text overlays that can also be changed per language.

I will need to determine which transitions will be used after each scene change and when characters enter and exit.

>> No.4145  

>>4137
This output does not have mid-line quotation marks escaped.

s10, line 310:

y "That aside, what happened to you being uncomfortable using "Onee-sama" while on campus?"

The hide statements are missing the image labels. hide tyu01s should be hide yu tyu01s. There are also some extraneous stop music and play music statements that simply restart the music that is already playing after a hide.

When a character enters or exits the scene, add with dissolve on the following line after the show statements.

show yu tyu01s at l
show na tna03s at r
with dissolve

After a combination of hide and show when a character exits, add at center to the show statement with the sole character remaining.

hide yu tyu01s 
show na tna01s at center
with dissolve
>> No.4146  

>>4145
This version has "at center" after multiple characters exit as well, is that correct?
http://www.mediafire.com/download/q8xl1x8i4rme5gf/JP_SH1_Remake_strings_v4.rar
http://www.mediafire.com/download/8db4e8ahvp1c0sb/EN_SH1_Original_strings_v4.rar

>There are also some extraneous stop music and play music statements that simply restart the music that is already playing after a hide.

Is that all music that starts playing when it's already playing? I commented those out, here they are:
(English, original)
strings_S002.MSD.txt: #Music already playing: M04
strings_S007.MSD.txt: #Music already playing: M01
strings_S010.MSD.txt: #Music already playing: M02
strings_S012.MSD.txt: #Music already playing: M02
strings_S013.MSD.txt: #Music already playing: M04
(Japanese, remake)
strings_S010.MSD.txt: #Music already playing: M02
strings_S012.MSD.txt: #Music already playing: M12
strings_S013.MSD.txt: #Music already playing: M09

>> No.4147  
File: 2014-09-02 11_15_57-A Kiss For The Pytals.png -(1069053 B, 808x627) Thumbnail displayed, click image for full size.
1069053

This is an attempt at a windowless text option. Unlike the Message Window Mod for the original visual novels, the appearance of the text can be changed for improved legibility.

>>4146
This version of the script adds a hide without any arguments where a cu frame should be hidden. It also hides the characters after the cu frame should be hidden, not before it is shown.

s002, line 299:

show yu tyu04s at l
show na tna03s at r
with dissolve

...

show cu02
#GS_HI_NS
stop music fadeout 1.0
play music "common/M04.ogg"

...

hide yu tyu04s
hide na tna03s
hide
show yu tyu05s at center
with dissolve

It also adds with dissolve before every instance of two characters on screen. I'd like to avoid that if possible.

s001, line 163:

show na tna04s at l
show yu tyu04s at r
with dissolve
#GS_CLR
voice "NANA0016"
n "Eh!? Ah, um, well......"

show na tna04s at l
show yu tyu02s at r
with dissolve
#GS_CLR
voice "YUNA0004"
y "Wait just a moment and let me mend that up for you."
>> No.4148  

>>4147
All right, I think I got it. So before "show cu0X" the characters get hidden.
http://www.mediafire.com/download/2mndrg7qyrbg5vx/EN_SH1_Original_strings_v5.rar

>> No.4149  

>>4148
Japanese: http://www.mediafire.com/download/f1v929sxsn36xqe/JP_SH1_Remake_strings_v5.rar

>> No.4150  

>>4148
Thanks to help from the developer of Ren'Py, I have a method to implement the built-in transitions used in the original visual novels.

For transitions other than GS_CLR, please add the transition to the scene statement as with GS_. Normal backgrounds generally open with GS_CUD and events with GS_EN_NS.

scene bg03a with GS_CUD
scene EV01 with GS_EN_NS

For cu overlays, they need to clear the entire screen with the scene already in place, which will properly remove the characters on screen and transition in the cu image at the same time. Rather than using hide statements, this requires a repeat of the previous scene statement, followed by show cu on the next line, and with the GS_ transition that follows it. The same is done when the cu image leaves the screen with an added with GS_ after the next show statements. Additionally, you will need to add with dissolve when the character(s) enter the screen for the first time after a scene.

For example, the block in s007

scene bg06a
#GS_CUD

...

show na tna01s
#GS_CLR
voice "NANA0442"
n "Onee-sama, will this spot be acceptable?"

...

hide yu tyu02s
hide na tna01s
show cu02
#GS_HI_NS

...

hide cu02
show na tna05s at center
with dissolve
#GS_CLR

should now read:

scene bg06a
with GS_CUD

...

show na tna01s with dissolve
#GS_CLR
voice "NANA0442"
n "Onee-sama, will this spot be acceptable?"

...

scene bg06a
show cu02
with GS_HI_NS

...

scene bg06a
show na tna05s at center
with GS_CLR

Note that there are not always any characters to show after a cu sequence ends.

>> No.4151  

I've added the project files to GitHub:
https://github.com/AXYPB/pytals

>> No.4153  
File: uniscript.txt -(5206 B, 0x0) Thumbnail displayed, click image for full size.

>>4150

>Normal backgrounds generally open with GS_CUD and events with GS_EN_NS.

All right. FYI these didn't have a GS_:
strings_S003.MSD.txt: scene bg08a
strings_S011.MSD.txt: scene ev12
strings_S015.MSD.txt: scene EV20B
strings_S015.MSD.txt: scene EV20

>Additionally, you will need to add with dissolve when the character(s) enter the screen for the first time after a scene.
>show na tna01s with dissolve

Should "with dissolve" be on a new line?
I think right now it's different between first appearance in a S0XX script and first appearance after a 'scene'. I'm not sure if that's correct.

http://www.mediafire.com/download/fm82x9q72k67gy0/EN_SH1_Original_strings_v6b.rar
http://www.mediafire.com/download/a7a9z5q56ewvdtz/JP_SH1_Remake_strings_v6b.rar

Current script attached.

>> No.4154  

>>4153
I've noticed some instances of a single character exiting the screen, leaving only the background. These are marked by a single GS_CLR statement with no characters immediately before or after it, but followed by a line of dialogue or narration.

s001, line 340:

show na tna03s at center
with dissolve
#GS_CLR
voice "NANA0029"
n "And that still was the only time that I had been able to speak directly with her..."

...

"To a middle-class girl like myself, the atmosphere here felt unusual indeed."

#GS_CLR
"Be that as it may, after two months in this environment, one does get used to it. And so, now..."

The GS_CLR here, surrounded by two lines of dialogue with no characters, should be replaced with scene bg03a with GS_CLR, as when a cu frame begins and ends. This will correctly have the characters exit the screen. When character(s) re-enter the scene, they will need to enter with dissolve as before.

Similarly, the cu frames end at the next GS_ transition, which is usually followed with character images. There are some points where this was not the case, which this script didn't account for.

s001, line 458:

scene bg03a
show cu01
with GS_HI_NS
voice "YUNA0017"
y "Come now, Nanami-san, you need to get undressed."

...

#GS_CLR
"Gyah?!"

At GS_ here, the cu frame needs to be cleared with the same scene bg03a that it followed, followed by with the GS_ transition that closes it. No characters will re-enter the screen. Again, they should enter with dissolve.

>Should "with dissolve" be on a new line?
>I think right now it's different between first appearance in a S0XX script and first appearance after a 'scene'. I'm not sure if that's correct.

When a with statement is on the same line as a show, scene, or hide statement, it affects only that statement. When it is on its own line, it affects all show, scene, and hide statements without their own with statements that preceded it simultaneously.
http://renpy.org/doc/html/displaying_images.html#with-statement

For only a single character, both styles are identical. I believe having the with statement on its own line in all cases where a character first enters a scene is safe.

For the exceptions you listed, in s003, I wasn't able to find scene bg08a without a GS_. The following three scenes are special cases with scrolling images. I will need to implement them separately.

I've made changes to fades in and out. After scene BLACK, please add with fade2 at the end of the line and not pause 1.0 on the next line. When the music changes, add stop music fadeout 1.0 on the line before the next scene, but only if there's a different music file after the next scene. The play music clause should come after the following scene and with lines.

The block in s003, line 104:

voice "NANA0168"
n "Hyaaaauuuu...."
scene BLACK
pause 1.0
scene bg08a
with GS_CUD
stop music fadeout 1.0
play music "common/M02.ogg"
show yu tyu02s at center
with dissolve

should now read:

voice "NANA0168"
n "Hyaaaauuuu...."
stop music fadeout 1.0
scene BLACK with fade2

scene bg08a
with GS_CUD
play music "common/M02.ogg"
show yu tyu02s at center
with dissolve

The block in s004, line 323:

scene bg03a
with GS_CUD
stop music fadeout 1.0
play music "common/M02.ogg"

should now read:

stop music fadeout 1.0

scene bg03a
with GS_CUD
play music "common/M02.ogg"

I will need to study the transform documentation to emulate the animation on the dialogue window between scene changes.

For convenience, in the future, please give your output files a .RPY extension and add label start: to the first line of s001.

>> No.4156  

>>4154
I have defined custom transitions for the message window. Immediately before every scene change, including EV and cu images, add window hide (except in the first scene at the top of the file), and immediately before the next line of dialogue or voice, add window show. Do not add these window commands between consecutive show EV commands, but add window hide before the first and last show EV and add window show on the next scene change after EV.

The block in s001, line 5:

scene bg02a
with GS_CUD
"It happened in April, as the sakura trees danced."

should now read:

scene bg02a
with GS_CUD
window show
"It happened in April, as the sakura trees danced."

s001, line 548:

"And so, bursting with joy, I had become a member of the committee."

scene bg03a
with GS_CUD
show yu tyu01s at center
with dissolve
#GS_CLR
voice "YUNA0020"
y "That's the last of the specifics for the changes to the final club activities budget."

should now read:

"And so, bursting with joy, I had become a member of the committee."

window hide
scene bg03a
with GS_CUD
show yu tyu01s at center
with dissolve
#GS_CLR
window show
voice "YUNA0020"
y "That's the last of the specifics for the changes to the final club activities budget."
>> No.4157  

>>4154
Some adjustments: For scene BLACK, please use stop music fadeout 2.0 instead of fadeout 1.0.

window hide
stop music fadeout 2.0
scene BLACK with fade2
scene bg03a
with GS_CUD
play music "common/M02.ogg"
show na tna03s at center
with dissolve
window show

For the first scene EV in a sequence, please also add pause 1.0 after each stop music fadeout 1.0 statement.

stop music fadeout 1.0
pause 1.0
window hide
scene EV01
with GS_EN_NS
>> No.4158  

>>4155

>Do not add these window commands between consecutive show EV commands, but add window hide windowout before the first and last show EV and add window show windowin on the next scene change after EV.

That should be 'scene EV', right? So after a 'scene EV' there's no 'window show windowin' and a 'scene EV' that goes into a another 'scene EV' has no 'window hide windowout' before the sub-scene?

>> No.4160  

>>4158
In the KiriKiri-based visual novels, the dialogue window is persistent between two EV images. I'd like to use that behavior in this port. Where there are more than one consecutive EV images, add window hide and window show only for the first and last EV images as with normal scene bg commands, but not from a scene EV to another scene EV. Another option is to change this to add window show and window hide between two differently numbered scene EV, but not for variants such as scene EV01p1. I'd like to see the former first.

Do not use windowin or windowout. I have instead defined them as default transitions.

>> No.4161  

>>4157

>For scene BLACK, please use stop music fadeout 2.0 instead of fadeout 1.0.

I think I got this, but do check both versions. Sometimes the Japanese version would do fadeout 2.0 even without a 'scene BLACK'. It checks for this when it's putting "stop music" commands before "scene" commands.
http://www.mediafire.com/download/e56oqhphzm8add2/JP_SH1_Remake_strings_v7.rar
http://www.mediafire.com/download/mt93rziietvq5ua/EN_SH1_Original_strings_v7.rar

>> No.4162  
File: 2014-09-04 19_18_45-A Kiss For The Pytals.png -(662506 B, 808x627) Thumbnail displayed, click image for full size.
662506

>>4161
I am using the Japanese script to compare image and music commands. I will allow those to be used instead if the user has a copy of a visual novel that uses them, not necessarily the remake of this one.

In the meantime, please adjust your script so that it doesn't replace ☆ with ♡; both are used in later titles. I have changed the font to "Tuffy", a free font that includes accented characters and the three symbol characters ♪☆♡. Please provide feedback.

>> No.4164  

>>4162
For branches, the code to replace sel01p and selc2_c.png is as follows:

menu:
"Choice 1":
jump b1a
"Choice 2":
jump b1b

label b1a:

Replace the number 1 with the number in the branch file name sel01_p. Because the MSD doesn't mark where each choice starts clearly, the labels for additional choices and the rejoin point will have to be inserted manually.

I plan to also use the same flag system from the original visual novels, where if the reader makes more than one incorrect choice, the story will prematurely end at the end of s012. However, an explanation of this will be added, and instead of returning to the title screen, the load game screen will be shown instead.

>> No.4165  

>>4164
The indentation for 'menu:' is 4 spaces, right? But for 'label b1a' should it be 0?

http://www.mediafire.com/download/84i0t99dmb3lchz/EN_SH1_Original_strings_v8.rar
http://www.mediafire.com/download/t5r0yd4k4oxldoi/JP_SH1_Remake_strings_v8.rar

>>4162
It looks good.

>> No.4166  

>>4165
This entire block is indented with four spaces. menu: and label b1a: are on the same indentation.

>> No.4167  

>>4166
But "label b1a:" and "voice "NANA0205"" can't have the same indentation, can they?

>> No.4168  

>>4167
Yes. The labels for branches do not need their own indented blocks. We would be able to add indentation for better readability if there were a mark for the end of the branch.

>> No.4170  

>>4168
All right, new version:
http://www.mediafire.com/download/boa8119bpfeqlbl/JP_SH1_Remake_strings_v8b.rar
http://www.mediafire.com/download/tz4pdj104so0emm/EN_SH1_Original_strings_v8b.rar

>> No.4173  

By the way in remake dialogues had defined pink color while narration had white. I saw that in files Yuuna and Nanami have defined which one is speaking so if color is not set all you need to write before 'label start' is:

define n = Character('Nanami', color="#fd62bb")
define y = Character('Yuuna', color="#fd62bb")

Or even different color for different character but you prabobly know all this.

>> No.4174  

>>4170
The end of each script needs to end with stop music fadeout 2.0.

window hide
stop music fadeout 2.0
scene BLACK with fade2
jump s002

I've also found an instance where a scene BLACK in the middle of a chapter also changes the music. The returning scene needs to be accompanied with stop music fadeout 2.0 as well.

s003, line 452, should read:

window hide
stop music fadeout 1.0
scene bg08a with fade2
play music "common/M02.ogg"
show na tna05s at center
with dissolve

I believe that the scripts are rapidly approaching a fully working state before manual adjustments are necessary. Please process the MSDs for My Dear Prince and Joined in Love with You after making the above corrections. Note that both of those visual novels also have a scrolling ev image that will need to be accounted for.

>> No.4175  

>>4174
Before the retranslation for the remake, I can try to take the Japanese remake script and substitute in the dialog from the original. Would that help?

>> No.4176  
File: 2014-09-05 18_42_38-A Kiss For The Pytals.png -(961693 B, 808x627) Thumbnail displayed, click image for full size.
961693

>>4173
The text colors can be easily changed at any time. We need not necessarily use the same colors from the original visual novel.

>> No.4177  

>>4175
If you can accomplish this with sed or some other automated method, by all means do so. I want to write in checks for the presence of image and music files from the remake to allow owners of the original version to use this framework, but also allow those who have the images from any visual novel that has the same images and music to use those instead, without any action needed from the user.

>> No.4179  

>>4175
Here are tab-delimited files containing the dialogue for both versions, if they will be of assistance.
http://www.mediafire.com/download/66lfsdl2dbx6h2q/SHP1.zip

>> No.4180  

>>4174

>Please process the MSDs for My Dear Prince and Joined in Love with You after making the above corrections.

Instead of many archives I can make one archive with subdirectories or everything in the same directory and named like SH2_EN_Remake_S001.rpy for instance. What do you prefer?

>> No.4181  

>>4180
Use subdirectories. I will need to create separate projects for each MSD. I am working with the first visual novel to create the common framework, which will be used as a template for all following visual novels.

>> No.4183  

>>4180
Please also add the following block to the start of s001, after label start:.

stop music fadeout 1.0
scene BLACK
with fade2

Here, with fade2 is on its own line. This will allow the main menu to fade to black after clicking on Start.

>> No.4196  

>>4180
I have found that the first line of dialogue after a scene EV is not being assigned a character according to the preceding voice file, but it does if the music also changes.

s001, line 878:

scene EV01
with GS_EN_NS
play music "common/M03.ogg"
window show
voice "YUNA0041"
y "Hurry... she's coming back... hurry... uhwaa..."

...

scene EV01p1
with GS_EN_NS
voice "YUNA0044"
"Uuuh, Nanami-chan... I love you, I love you! Nanami-chaaaan!"
>> No.4197  

>>4174

>I've also found an instance where a scene BLACK in the middle of a chapter also changes the music. The returning scene needs to be accompanied with stop music fadeout 2.0 as well.

What do you mean? Can you give an example?

So far the script seems to be working well with the other games. SH2 S006 has a Japanese savegame title which I changed, あなたしか見えない - I Only Have Eyes For You.
http://www.mediafire.com/download/fa6atkxx8hg4kgt/SH123_strings.rar

>>4196
I will look into it.

>> No.4198  

>>4197
And the character codes I used are these:
#game 1
s/YUNA "/y "/
s/NANA "/n "/
s/KYOS "/tt "/

#game 2
s/KAED "/k "/
s/SARA "/s "/
s/KYS. "/te "/
s/KAEH "/km "/

#game 3
s/MAI. "/m "/
s/REO. "/r "/

#recurring
s/GALA "/ga "/
s/GALB "/gb "/
s/GALC "/gc "/
s/JOSA "/ga "/
s/JOSB "/gb "/
s/JOSC "/gc "/
s/JOSD "/gd "/
s/JOSE "/ge "/
s/JOSF "/gf "/
s/JOSG "/gg "/

>> No.4199  

>>4196
http://www.mediafire.com/download/egv96qrlsfm052b/SH123_strings_v2.rar

>> No.4201  

>>4197
The chapter title I have for s006 is "You're the Only One I See". Was it not present in the current translated MSDs?

>>4198
If you have a GitHub account, I'd like to add you as a collaborator for the repository at http://github.com/AXYPB/pytals so that you can upload your bash scripts.

>> No.4202  

>>4198
Joined in Love with You has two other voiced characters:

  • Mai's Mother - maih
  • Mai's Sister - maim

There is also Mai's Brother, but he is unvoiced.

My Dear Prince has a character "Girls" for all of the extra girls exclaiming in unison, but I'm unsure what its voice file prefix is.

>> No.4205  

>>4201

>The chapter title I have for s006 is "You're the Only One I See". Was it not present in the current translated MSDs?

The one I got was from the prepatched archive though. Are they outdated? If so can you upload the new MSDs?

>If you have a GitHub account

Yes, I have one: Procyon2

>>4202
Okay, got it.

And I discovered the pattern that says who the character is. It is 45 bytes long and ends 12 bytes before the 0x03 of the voice string.

There are four bytes (denoted below with !) that change, the first three are 0xffffff if it's a minor character, unknown character or the narrator and the last byte is the character number (0xff for narrator).

Example:
SH3, unknown character:
#01 FF!00 00 00 01 FF!00 00 00 01 FF!
#00 00 00 D1 07 14 00 01 01 00 00 00
#01 00 00 00 00 01 00 00 00 00 01 00
#00 00 00 D7 07 05 00 01 0F!

SH3, Reo:
#01 D7!00 00 00 01 EF!00 00 00 01 E0!
#00 00 00 D1 07 14 00 01 01 00 00 00
#01 00 00 00 00 01 00 00 00 00 01 00
#00 00 00 D7 07 05 00 01 01!

Here are all the character codes (the last byte):
Game 1
00: CODE:-NANA:
01: CODE:-YUUNA:
02: CODE:-GALA:
03: CODE:-GALB:
04: CODE:-GALC:
05: CODE:-KYOS:
06: CODE:-CHILDA: #no voice
07: CODE:-CHILDB: #no voice
08: CODE:-UNKCHAR:

Game 2
00 CODE:-KAEDE:
01 CODE:-SARA:
02 CODE:-YUUNA:
03 CODE:-KAEH:
04 CODE:-KAEDE_FAMILY: #no voice, probably the father
05 CODE:-JOSA:
06 CODE:-JOSB:
07 CODE:-JOSC:
08 CODE:-JOSD:
09 CODE:-JOSE:
0a CODE:-JOSF:
0b CODE:-JOSG:
0c CODE:-UNUSED?:
0d CODE:-KYS:
0e CODE:-UNKCHAR:

Game 3
00 CODE:-MAI:
01 CODE:-REO:
02 CODE:-YUUNA:
03 CODE:-KAEDE:
04 CODE:-KYOS:
05 CODE:-JOSA:
06 CODE:-JOSB:
07 CODE:-JOSC:
08 CODE:-JOSD:
09 CODE:-JOSE:
0a CODE:-JOSF:
0b CODE:-JOSG:
0c CODE:-MAIM:
0d CODE:-MAIH:
0e CODE:-MAI_BROTHER: #no voice
0f CODE:-UNKCHAR:

I'll add the unvoiced and unknown characters like this:
UNKCHAR - q
CHILDA B - ca cb
KAEDE_FAMILY - kf
MAI_BROTHER - mb

>My Dear Prince has a character "Girls" for all of the extra girls exclaiming in unison, but I'm unsure what its voice file prefix is.

I didn't see it. If you know the line let me know.

>> No.4206  

>>4205
I've added you to the repository. You can add your scripts to the root. I will create folders for the other visual novels as well.

Good work on finding the indicator for the name assignments. These are the additional characters for all other MScenario visual novels:

Beloved Photograph:

  • Newspaper Reporter (news)
  • Woman A (woma)
  • Woman B (womb)
  • Clerk (tenn)
  • Photographer (unvoiced)
  • Staff (unvoiced)

The Joy of Loving You:

  • Class Representative (icho)
  • Clerk (tenn)

Whisper With a Kiss:

  • Crowd (sunn)

Sweet Enchanted Kisses:

  • Vice Principal (Kyotou)
  • Girls A-C (Mix)

The following visual novels define the extra girls as JyosiA, JyosiB, etc.

Dyed With an Angel's Petals:

  • Girls A-C (Mix)
  • Homeroom Teacher (Tannin)
  • Maid (Otetud)
  • Nurse (Hokeni)

Sweet Grown-up Kisses:

  • Girls A-C (Mix)

Lily Platinum:

  • Girls C & D (JMixCD)
  • Club President A (ButyoA)
  • Club President B (ButyoB)
  • Club Member A (Buin_A)
  • Club Member B (Buin_B)
  • Club Member C (Buin_C)
  • Club Member D (Buin_D)
  • Club Members C & D (BMixCD)
  • Homeroom Teacher (Tannin)
  • Maid (Otetud)

There are also two other extra girl characters 女の子A (OnnaA) and 女の子B (OnnaB) that are distinct from the girls in the school. I believe they appear at s012. We will need a different label for them.

Maidens of Michael:

  • Risa's Mother (RiHaha)
  • Kaede's Mother (Kahaha)
  • Manager (Manage)
  • Teacher (Kyosi)
  • Committee Member (Iin)
  • Committee Member A (IinA)
  • Committee Member B (IinB)
  • Committee Member C (IinC)
  • Reporter (Kisya)
  • Staff (Stuff/stuff)
  • Driver (Unten)

I'm unsure what these characters are at this time as the translation has not yet reached their first appearances.
* 司会者 (Sikai; appears during Kaede/Sara route)

  • 社長 (Syatyo; appears during Kaede/Sara route)
  • 案内 (Annai)

There are also several voice files labeled "Mix", but they are not specific as to which characters are mixed. The name image file defines mixes for Eris and Reo, Shizuku and Kaede, and Mai and Kaede. There is also a name 全員 for all characters, which may be what the Mix is referring to. If this is the case, the mixed lines may be named the same as the characters in question, but have a different byte marker than normal in the MSD.

My Sworn Love for You:

  • All (All)
  • Woman A (KyakuA)
  • Woman B (KyakuB)
  • Woman C (KyakuC)
  • Woman D (KyakuD)
  • Woman E (KyakuE)
  • Woman F (KyakuF)

The woman characters are named 女性客 female visitor, but I believe this can be shortened.

>> No.4207  

>>4206
I'll just run the script inside the repository directory, right? Should I add the S00X.MSD files, too? It would help if you can put the rest of those up somewhere.

Can I give those characters a 2-letter character code myself like above (ca, cb, etc)?

>> No.4209  

>>4207
I added them, but maybe they go better in a 'build' subdirectory.

>> No.4210  

>>4207
It may be better, if slightly less efficient, if we redefine the characters with their full names, with the secondary characters abbreviated.

For the MSD files, you'll only need the s0XX files. The others can be excluded.

I created a new branch in the repository, new-processed. Upload the output of new versions of your bash script to this branch. I will make manual edits to it and merge it into master.

Here are the latest MSD files for My Dear Prince.
http://www.mediafire.com/download/bapmhcd2rmtywa9/MSD2.zip

>> No.4211  

>>4210
I changed the dialog character codes to full names. To make it easier to identify programmatically I gave each an underscore.

>>4174

>I've also found an instance where a scene BLACK in the middle of a chapter also changes the music. The returning scene needs to be accompanied with stop music fadeout 2.0 as well.

I still don't know what you mean with this.

I noticed that right before a "show cu" the previous scene is displayed again to clear all the characters, but what if it's "scene BLACK" (I noticed that that could happen quite a bit further back in the script)? The latest commit has without "scene BLACK" before "show cu" instead showing the scene before it and gives a warning. Could you see what it's supposed to do there?
1_English_Original/strings_S001.MSD.rpy:486:
1_English_Original/strings_S003.MSD.rpy:332:
1_English_Original/strings_S003.MSD.rpy:380:
1_English_Remake/strings_S001.MSD.rpy:486:
1_English_Remake/strings_S003.MSD.rpy:332:
1_English_Remake/strings_S003.MSD.rpy:380:
1_Japanese_Remake/strings_S001.MSD.rpy:486:
1_Japanese_Remake/strings_S003.MSD.rpy:332:
1_Japanese_Remake/strings_S003.MSD.rpy:380:

2_English_Remake/strings_S002.MSD.rpy:792:
2_English_Remake/strings_S002.MSD.rpy:942:
2_English_Remake/strings_S010.MSD.rpy:220:
2_English_Remake/strings_S010.MSD.rpy:273:

3_English_Remake/strings_s001.MSD.rpy:167:
3_English_Remake/strings_s001.MSD.rpy:194:
3_English_Remake/strings_s001.MSD.rpy:1033:
3_English_Remake/strings_s006.MSD.rpy:361:

>> No.4212  

>>4211
In 1_English_Original/strings_S003.MSD.RPY, there is a point where the scene fades to black and the music changes. When this short segment ends, the previous music and scene resumes, but there is no fadeout statement to return to the previous music, so the current music cuts immediately instead of fading out.

Your current scripts have corrected the stop music fadeout 1.0 issue, but they do not repeat the scene before BLACK. To my knowledge, cu images are never preceded by scene BLACK, but I haven't checked all scripts thoroughly.

A Kiss For The Petals s003, line 454:

window hide
stop music fadeout 1.0
play music "common/M02.ogg"
show na tna05s at center
with dissolve
#GS_CLR
window show
voice "NANA0198"
nanami_oda "Kyaa-kyaa!! Nonononoo♪ Something like that would {i}kill{/i} me!♪"

A second scene bg08a (from line 381) is needed here after stop music fadeout 1.0.

In places where 『』 are used for dialogue from off-screen, your scripts have inserted two quotation marks at the start and end of each line. There should be only one set of quotation marks even for these lines.

Joined in Love with You s006, line 122:

voice "REO0362"
reo_kawamura ""H-Hello... It's me.""
>> No.4215  

>>4212
There is also at least one line of narration in parentheses. These will need to be enclosed in quotation marks.

My Dear Prince s001, line 85:

(There aren't any students in St. Michael's who hold petty thoughts like that.)

Additionally, in the remake scripts, file name case is mixed. There are references to lowercase bg and uppercase BG. These will need to be accounted for.

>> No.4216  

>>4210
The original has an English title for SH2 S006, but in the remake it's Japanese. I'll just use the remake and change only the title.

>>4215

>Additionally, in the remake scripts, file name case is mixed. There are references to lowercase bg and uppercase BG. These will need to be accounted for.

Should I change references in the S00X.rpy scripts to match define.rpy?

>> No.4217  

>>4216
Yes, we will use the file names in their original cases to avoid problems with extraction.

I will also need to issue a fix for the translation patch of the remake version of My Dear Prince. This missing chapter title has never been reported.

>> No.4218  

>>4217

>Yes, we will use the file names in their original cases to avoid problems with extraction.

I didn't see any mismatches.

>> No.4219  

>>4218
The remake scripts refer to bg18a and above, but the files themselves are named BG18a in MGD. This results in "image missing" errors in Ren'Py.

>> No.4220  

>>4218
Please see the define.rpy files for the respective visual novels for the exact image file names.

>> No.4223  

>>4220
All right, I put a better check into the script and now I see these mismatches (and they are auto-corrected):
game 2 S001 script: bg18a vs define: BG18a
game 2 S003 script: bg18a vs define: BG18a
game 2 S010 script: bg18a vs define: BG18a
game 2 S010 script: bg18b vs define: BG18b
game 2 S013 script: bg25a vs define: BG25a
game 2 S015 script: bg18a vs define: BG18a
game 3 s001 script: bg18a vs define: BG18a
game 3 s001 script: bg19a vs define: BG19a
game 3 s001 script: bg20a vs define: BG20a
game 3 s002 script: bg21a vs define: BG21a
game 3 s002 script: bg21a vs define: BG21a
game 3 s003 script: bg21a vs define: BG21a
That's checking both "show" commands and "scene" commands. (bgXX falling under scene)

>> No.4224  

>>4223
If you haven't already, you may need to write the same checks for the voice files (see >>4206).

>> No.4225  

>>4223
Your latest commit appears to be very close to the original behavior. Please run your script on all visual novels up to Maidens of Michael, using translations where available. For Lily Platinum, you can use this MSD linked in the translation thread: http://www.mediafire.com/download/fw0yjjtkaggsf7c/MSD_temp_fix.rar

I believe Maidens of Michael will present its own challenges.

>> No.4227  

>>4224
Should I change voice commands to this form:
voice "common/josa0001.ogg"

>> No.4228  

>>4226
Voice files do not need to be defined this way in Ren'Py. There is a global variable config.voice_filename_format = "common/{filename}.ogg" for this.

To reduce script file size, I believe we can shorten the character name definitions to given names only. We can also change "unknown_questionmarks" to "unknown".

>> No.4230  

>>4228

>shorten the character name definitions to given names only

Yeah, sure.
Game 7 has a WKYOSI, should I group that and both teacher_kyos taecher_kys as just "teacher"?
In the bash script I have the names as "name@" but I'm removing the @ when I put it in the repository.

>There are also several voice files labeled "Mix", but they are not specific as to which characters are mixed. The name image file defines mixes for Eris and Reo, Shizuku and Kaede, and Mai and Kaede. There is also a name 全員 for all characters, which may be what the Mix is referring to. If this is the case, the mixed lines may be named the same as the characters in question, but have a different byte marker than normal in the MSD.

In game 7 and 9, there are two hex codes for MIX, so if you want to change the character code, let me know which line/voice command.

In game 10 there are two hex codes for BUIN_A, which is probably irrelevant.

Games 5, 7, 8, 9 and 10 don't seem to have an unknown "???" character.

In game 8, who is JYOKYO? Her only line is "Okay, time's up."

>> No.4231  

>>4230
The heart emote is giving me some difficulties. Is the hex code something like 0x000164000000?
That hex code shows up before voice strings, too, giving lines like "♡RUNA0185"

>> No.4232  

>>4230
Can you give the lines of dialogue where these issues occur?

>> No.4233  

>>4231
As I mentioned in >>4076, the heart symbol in hex appears to be 00 01 64 00 00 00 03.

>> No.4234  

>>4232

>Can you give the lines of dialogue where these issues occur?

The heart emote before voice lines?

>> No.4235  

>>4234
I need to see the files that contain the hex codes you mentioned and the dialogue next to them.

>> No.4236  

>>4235
I turned the hearts off in my last commit so you can see it in a diff here (red is with heart emotes, green is without):
https://github.com/AXYPB/pytals/commit/7fd916c2c6325684c15320746b0d63c6ec6756bd

Example:
10_English/strings_S006.MSD.rpy

  • "♡SIZUKU0286"
  • "A flower resembling the lotus, which closes in the evening. So it is known as the Sleeping Lotus. That's what the water lily means."
  • voice "SIZUKU0286"
  • shizuku "A flower resembling the lotus, which closes in the evening. So it is known as the Sleeping Lotus. That's what the water lily means."

10_English/strings_S010.MSD.rpy

  • "♡ERISU_0410"
  • "It'll be fine. After all, I'm here. ♪Shizuku can rest at ease."
  • voice "ERISU_0410"
  • eris "It'll be fine. After all, I'm here. ♪Shizuku can rest at ease."
>> No.4237  

>>4236
I think I can tell them apart because the wrong ones are preceded by another hex code XX07XX (or perhaps others that should be identifiable).

>> No.4239  

I will be traveling starting tonight and returning early Monday morning, so my availability until next week may be limited.

>>4236
It appears that when you omit the heart symbol, your script is converting them into newlines. Can you explain why 1_English_Remake in this commit are mixing English and Japanese dialogue? I'm unsure what else to make of this commit.

>> No.4241  

>>4239
Both of those have been fixed since >>4237, right?

>> No.4242  

>>4241
It appears that they are.

If you have the time, can you determine the hex codes for the branches and where the choices jump to in the script? They begin near sel02_c. If possible, also check for code that marks which choices are "correct" and add to the counter that determines whether the story continues after the check at s012 or s015.

Beloved Photograph s003 has a branch with three choices after the line "Yup. I wonder what you should call me.", and there are three points where the dialogue changes depending on the choice made. Maidens of Michael has a branch at s024 with six choices, which determine the route after s026.

>> No.4249  

>>4242
I've only had partial success. I found a "Correct choice" (or something like it) appear in one of the branches for all choices (except SH4 S003 which has no bad answers). But, apart from a few exceptions, that's it. As far as I can see there's no reliable way to see where a branch starts or ends. There are a few exceptions to this though, like for SH4 S003 the branches have an indication at the start and the later few divergences as well, but still no convergence hex code, or anything like an absolute or relative line number that says where to jump to. One of the codes that indicated convergence in a few cases also showed up in completely different parts, but I'll leave them in. I also didn't see anything at the "Halfway Ends".

I did find the hex code for line number and realized it can be used to verify whether all lines are quoted. I fixed a few cases that way and a few commands that were interpreted as text. I saw three lines in Sono Hanabira 10 that had poor punctuation which I guess should be fixed in the official version, too:

SH10 S016
i continually felt this anxiety while on the campus
SH10 S017
only myself had been cast into a different flow of time. that's how i felt
, Eris' hips shook in response.

>>4135

>This may become further complicated in scenes in later visual novels with three characters on-screen.

I haven't seen three characters in a row. Do you know where this occurs, exactly?

>> No.4250  
File: 2014-09-14 10_52_30-A Kiss For The Petals - My Dear Prince.png -(763487 B, 806x625) Thumbnail displayed, click image for full size.
763487

>>4249
If this is the case, then we will have no choice other than to manually record the branch data.

The first instance of three characters at once is in My Dear Prince s008. The character codes begin at 0x22BC and are tyu02s tka04s tsa04s.

>> No.4253  
File: example_branches.txt -(91475 B, 0x0) Thumbnail displayed, click image for full size.

>>4250
Is 'at m' okay for that?

>show yu tyu02s at l
>show ka tka04s at m
>show sa tsa04s at r

When it goes from 2 characters at once to 3 characters at once, does it need a dissolve or something?

About branches, I am trying out adding in the labels and jumps for them using exact text lines. I'm trying it out for game 1 like this:
game1s003_correct=2
game1s003_branch_b="Uuh... I don't hate it." (branch b starts with that sentence)
game1s003_converge="Well then, would you say it for me right now please?" (the first sentence after convergence)

Could you look at this example file? (search for 'label') It uses score+=1 at the branches I got from a walkthrough. It's all of game 1's files with branches in them, just to see if it worked.

>> No.4254  
File: choices.txt -(3230 B, 0x0) Thumbnail displayed, click image for full size.

I autogenerated it with a script that uses data from this file. It would need to be filled in. SH4 S003 is just the first choice. The later few branches may as well do manually. They do get marked with comments.

>> No.4258  

>>4253
Variable assignments are Python code and must be preceded by $.

$ score += 1

The score addition should be called at the start of the label for the correct choice, not at the end of the branch. The script you attached also added a @ to every name label.

For scenes with three characters, add the transition as with other cases where the number of characters on-screen changes. I will also need to define new transforms for these cases, at l2 and at r2. Use at center as normal for the second character.

>> No.4260  

>>4258
I'm only loosely following this discussion, but I thought I might point out there's at least one instance in Maidens of Michael where there are four characters on screen simultaneously. I don't know whether this would need to be handled any differently from scenes with three characters, but I thought it might be worth mentioning.

>> No.4262  

>>4260
Can you post a screenshot, the point in the script where it occurs, and a save file?

>> No.4264  
File: Capture.JPG -(110602 B, 806x629) Thumbnail displayed, click image for full size.
110602

>>4262
This occurs in s012.
The line in the script is:

  • <99CE> 「ま、待ってくださ~い、はぅぅ~」
>> No.4265  
File: GAME49.txt -(214862 B, 0x0) Thumbnail displayed, click image for full size.

>>4264
Here is the save file.

>> No.4272  

>>4258

>The score addition should be called at the start of the label for the correct choice, not at the end of the branch.

It's functionally the same.

>>4254
Are you filling these in? I could take some games if you're too busy.

>> No.4273  
File: 2014-09-21 12_50_30-A Kiss For The Petals - Whisper With A Kiss.png -(713253 B, 806x625) Thumbnail displayed, click image for full size.
713253

>>4272
I've added all branch points to choices.txt in the repository. There are also a number of special effects in the later visual novels that require individual attention. I will address them when the text parsing process is finalized. I've added #TODO comments to the ones I encountered while doing this.

For the convergence points you added comments to, they appear to mostly coincide with scene changes, but there is no consistency to this.

>>4253
For scenes with three characters, continue using at center. It isn't necessary to add another transform.

I've also found that there are some cases where cu images directly follow scene ev images. In these cases, the cu image is shown over black.
My Dear Prince s010, line 221:

scene BLACK with fade2
scene EV004p1
show cu03

Line 274:

scene EV004p1
show cu03P1
with GS_R

Where a show cu statement follows scene BLACK, the previous scene EV statement needs to be removed. The preceding scene in this case is scene BLACK.

>> No.4274  
File: 2014-09-21 14_31_31-A Kiss For The Petals - Dyed With An Angel's Petals.png -(279051 B, 806x625) Thumbnail displayed, click image for full size.
279051

Dyed With an Angel's Petals uses an unusual distortion transition two times in s014. The below save file begins immediately before it.
http://www.mediafire.com/download/kldqq5bgqk2uysn/GAME14.zip

>> No.4276  

>>4273

>I've added all branch points to choices.txt in the repository.

It's working well. There's a typo in the original script by the way:
voice "SARA0157"
"Please lick my naughty little pussy your tongue, Ojou-sama..."
I'll fix it on this side.

The SH4 S003 branches will now set a variable.
In the main branches, for example Choice 1:
$ sh4s3 = 1
And the subbranches will check it like this:
$ if sh4s3 == 1: jump subbranch2a
$ if sh4s3 == 2: jump subbranch2b
$ if sh4s3 == 3: jump subbranch2c
This is mixing python and renpy code. Will it work like that?

>There are also a number of special effects in the later visual novels that require individual attention. I will address them when the text parsing process is finalized. I've added #TODO comments to the ones I encountered while doing this.

You should realize the strings files are recreated from scratch each time. If it's not in the script creating the strings files, changes will be lost. They can still be viewed in github's diff of course:
https://github.com/AXYPB/pytals/commit/6fff1d6ed791f80c2637f416ae2c156a68052eb0

In SH7 S006 you changed the save game name:
$ save_name = "Traces of \"Her\""
There are no quotes in the MSD file.

>> No.4277  

>>4276
I believe this code will work. Alternatively, a python: block can be used for multiple lines of Python code.
http://renpy.org/doc/html/python.html

In choices.txt, is it necessary for quotes within lines to be escaped with \\\?

Let us make corrections to the script in these files. After these scripts are released as Ren'Py packages, I plan on giving priority support for them over the original visual novels.

>In SH7 S006 you changed the save game name:
>$ save_name = "Traces of \"Her\""
>There are no quotes in the MSD file.

I have a set of save files for this visual novel at every chapter start that were used for debugging. They may contain outdated save titles. It is possible to use formatting on save titles in Ren'Py, as in "Traces of Her".

>> No.4282  

>>4274
I found the hex code for that transition. But after highlighting similar codes, I only saw transitions already handled by a GS.
>>4277

>It is possible to use formatting on save titles in Ren'Py, as in "Traces of /Her/".

I changed it to "Traces of {i}Her{/i}", is that correct?

>> No.4283  

>>4282
Yes, I used the same change in my own script, but didn't commit it.

Can you find the hex codes for all transitions and their corresponding plaintext names?

>> No.4284  

Sono Hanabira 4 S015 has a command called "WHITE":
4_English/strings_S015.MSD.rpy-181- voice "KAED1030"
4_English/strings_S015.MSD.rpy-182- kaede "And... Go!"
4_English/strings_S015.MSD.rpy-183-
4_English/strings_S015.MSD.rpy-184- WHITE
4_English/strings_S015.MSD.rpy-185- voice "BEV020"
4_English/strings_S015.MSD.rpy:186: xpause 1.0xstop music fadeout 1.0
4_English/strings_S015.MSD.rpy-187- play music "common/M05.ogg"
4_English/strings_S015.MSD.rpy-188- voice "SARA1201"
4_English/strings_S015.MSD.rpy-189- sara "Kaede-chan, you're so beautiful!♪"

Should I turn this into "scene WHITE"?
It's followed by twice BEV020 (the repeating one was removed automatically) which is a scene, I guess?
The "xpause 1.0x" is from a related error.

>>4283

>Can you find the hex codes for all transitions and their corresponding plaintext names?

It's not going well. There's no plaintext left in any case, the strings program caught it all. If you have more non-standard transitions I could look into that. I added a few things that stood out to me with "#TYPE:<BYTE>" but I'm not sure if it's at all from a useful location.

>> No.4288  
File: BEV020.png -(1962303 B, 1600x1200) Thumbnail displayed, click image for full size.
1962303

>>4284
"BEV020" is an enlarged version of "EV020" which scrolls from bottom to top, as with the tall "ev" images from the first three visual novels.

What are the -NULL: lines that were added through many of the scripts in the latest commit? The TYPE lines will all need to be commented out.

>> No.4298  

>>4288
I'm trying out a new method that starts out with hex (instead of only strings) which should make it faster to check certain sections of the MSD file for unique patterns.
It shows strings, known binary commands as names with binary arguments, and unknown binary commands/arguments as hex.

But it still needs a lot of work and it'd be really hard to find new patterns, only other commands that are verified.

I looked through the TODOs that were in the above commit and I at least found more instances of these two:
#TODO: This scene uses transparencies. The background is called again with a transition and darkened, and Yuuna is transparent and Nanami is opaque.
#TODO: Here, tru04s is hidden, but tta01s remains in place; it does not move to center.

For example, when a character stays on the left instead of moving to the center:
(B: is binary, S: is string. Some command names are wrong, but whatever. image_related_3 is image placement)
B:(pre_gs ecffffff type) (image_related_3 04) (image_related_3 05) (image 64 04)
S:tru03s
B:00 64 00 09 00 01 05 00 00 00 03
S:NO

And on the right:
B:(pre_gs ecffffff type) (image_related_3 04) (image_related_3 05) 00 00 00 64 00 09 00 01 04 00 00 00 03
S:NO
B:(image 64 05)
S:tta02f

I can now add that to the main program. It would come out like this, right?
show tru03s at l
with dissolve

For transparency should it look like this?
show tyu02s
with transparency

The hex code for transparency was 00 00 69 00 19 00 01 etc:
B:(unit start) (image_related_2) (bin_music)
S:M11
B:(postmusic) 00 00 69 00 19 00 01 00 00 00 00 01 ff 00 00 (post_gs_clr) (post_gs_clr) (post_gs_clr) 00 69 00 19 00 01 04 00 00 (post_gs_clr) 00 01 ff 00 00 00 01 ff 00 00 00 01 ff 00 (predammy)
S:DAMMY
B:(pre_gs ecffffff type) (image_related_3 04) (image 64 04)
S:tyu02s
B:(unknown_1 660404) (pre_gs 04c8c9ca) (gs_related_1 1c bc02)
S:GS_CLR
B:(post_gs_clr) (unknown_3 ffffff) (line 0c00) (prechar) (character:ffdfbd01) (postchar 0f 1800 type 3*00)
S:YUNA0213
B:(preline 3a 1900)
S:「What a great idea, Nanami♪」

>> No.4301  

>>4298

>show tru03s at l
>with dissolve

When a single character replaces multiple characters and is not centered, the previous character images need to be hidden first.

>show tyu02s
>with transparency

According to the documentation, transforms such as alphas can be applied to images by adding a colon to scene and show statements, which invokes ATL.

scene bg08a:
alpha 0.5
show tyu02s:
alpha 0.5

http://www.renpy.org/doc/html/atl.html

The most recent commit shows all of your #TYPE: comments deleted. Is this correct?

>> No.4302  

>>4301

>When a single character replaces multiple characters and is not centered, the previous character images need to be hidden first.

Hide all and reshow the remaining?
hide tru04s
hide tta01s
show tta01s at r
with dissolve

>The most recent commit shows all of your #TYPE: comments deleted. Is this correct?

Yes, that method was too ineffective.

>> No.4303  

>>4302
The remaining character(s) do not need to be hidden.

>> No.4306  

I added both the transparency and characters staying, but there's no hex code for turning transparency off, so I wonder if it's not staying on too long.

>> No.4308  

I think I found the hex for this one:
#TODO: tsa02s slides in from right to center, not with a dissolve. The sound SE033 plays while the image moves.

It also happens two or three times in SH4, but without SE033, only SE047. In SH5 SE033 (footsteps, 3.371 seconds) plays first and then SE047 (sliding shoes, 0.92 seconds) plays while the image moves.

SH4 S001:
"At that same moment, there was the sound of someone running in the halls."
hide ka tka01s
show sa tsa02s at center
with dissolve
play sound "common/SE047.ogg"
voice "SARA0012"
sara "Kaede-chan, let's go home together~"

In SH5 Sara says "Kaede-chan" while moving, but in SH4 the sound finishes playing. It's probably nicer to do it like SH5. The documentation seems to say to do this by appending channel 0/1.

hide ka tka01s
play sound "common/SE047.ogg" channel 0
voice "SARA0012" channel 1
show sa tsa02s:
\ xcenter 1.1
\ linear 0.92 xcenter 0.5
sara "Kaede-chan, let's go home together~"

SH5 S003:
"Just then, I heard footsteps down the hallway."
play sound "common/SE033.ogg"
hide ka tka02s
hide ma tma01s
show sa tsa02s at center
with dissolve
play sound "common/SE047.ogg"
voice "SARA0001"
sara "Kaede-cha~n!♪"

To:
hide ka tka02s
hide ma tma01s
play sound "common/SE033.ogg" channel 0
play sound "common/SE047.ogg" channel 0
voice "SARA0001" channel 1
show sa tsa02s:
\ xcenter 1.1
\ linear 0.92 xcenter 0.5
sara "Kaede-cha~n!♪"

And what should I do with the screen distortion? Something like this?
window hide
scene bg20 with distortion
window show
"Gradually, my vision returned."

>> No.4311  

The stop music fadeout 2.0 seems to be inserted unnecessarily between scene changes.
Beloved Photograph s001, line 60:

window hide
stop music fadeout 2.0
scene BLACK with fade2
scene bg03a

The music stops here when it should continue playing.

The branch in Beloved Photograph s003 does not need $ before its checks.

if sh4s3 == 1:
jump subbranch2a
if sh4s3 == 2:
jump subbranch2b
if sh4s3 == 3:
jump subbranch2c

The third sub-branch needs to come after scene EV005p1, not before.

scene EV005p1
with GS_EN_NS

if sh4s3 == 1:
jump subbranch3a
if sh4s3 == 2:
jump subbranch3b
if sh4s3 == 3:
jump subbranch3c

label subbranch3a:

The eyecatches need to be commented out as well, as in #TODO: eyecatch01.

>> No.4327  

>>4311
I noticed that a cu may end before a scene change. There's a hex code for it. It hides the text box, cu and characters and then reshows the text box.

>The stop music fadeout 2.0 seems to be inserted unnecessarily between scene changes.

I'm not sure why I added it. Is it okay now for all changed lines?

>> No.4343  

>>4327
After a cursory check, the music stop issue appears to be resolved.

I tested your clearscreen code in Sweet Grown-up Kisses and found a case in s001 where a CU image transitions into another (CU01 transitions to CU01p1) with a different music track. The music needs to change before scene BG26c.
s001, line 83:

play sound "common/SE020.ogg"
scene BG26c
show CU01p1
with GS_R
stop music fadeout 1.0
play music "common/M18.ogg"
voice "TAKAKO0000"
takako "Hold it~!"

I also found that some BG and 'CU' image file names are uppercase (BG22a, BG23a, BG23b, BG24a, BG25a, BG25b, BG26a, BG26c, CU01, CU02, CU03), but your script inserts them all as lower case. The cases are corrected in the code above. s004 references a scene "_EV05", which your script does not parse. The heart character is also translated as ♡, rather than ♡.

This may be an issue with the translated MSD, but an invalid character is inserted in s011.
s011, line 610:

"The things she does are so cute, there"fs no way I could pick the top 10.
>> No.4347  

>>4343
I noticed that the translation of s011 in the wiki uses a fullwidth apostrophe in the line in question, which is causing the error there. I've updated the S011.MSD file in the repository.

<B5C1> 全てが可愛いから、きゅんとくる瑠奈の仕草ベスト10なんて、とても決められない。=The things she does are so cute, there’s no way I could pick the top 10.
>> No.4431  

I've created structure data for the first ten visual novels in the repository. It should now be possible to run them in the Ren'Py SDK by checking out the appropriate files and extracting the corresponding data.

>> No.4435  

>>4343

>The heart character is also translated as ♡, rather than ♡.

Is this still a thing? Should the encoding of the script files not be UTF8?

Were you able to do a thorough investigation of the renpy scripts? I wonder if there any big bugs left.

>> No.4439  

>>4435
After reviewing the character name images, I noticed several characters weren't present in character_codes.txt.

7_English\define.rpy:

#TODO: There are two types of mix: Girls A & B and Girls A-C. 

10_English\define.rpy:

#TODO: There are also two other extra girl characters 女の子A (OnnaA) and 女の子B (OnnaB) that are distinct from the girls in the school. I believe they appear at s012. We will need a different label for them.

I haven't been able to audit the code in detail, and I have reason to believe there are still several issues left in the code. The scripts are nominally UTF-8, and I'm unsure what would cause the heart character to be corrupted in the conversion.

I would like to take the opportunity to ask about attribution. My current plan is to give you equal authorship credit with myself. Do you have any objections to this?

>> No.4450  

>>4439

>My current plan is to give you equal authorship credit with myself. Do you have any objections to this?

No, not at all.

>#TODO: There are two types of mix: Girls A & B and Girls A-C.

This is just a few lines per game, so you could change the strings file manually (I can't tell which girls mix is which).
SH7:

>7_English/strings_s004.MSD.rpy: voice "MIX0000"
>7_English/strings_s004.MSD.rpy- mix " “Kyaaaaaaaah!!”"
>7_English/strings_s005.MSD.rpy: voice "MIX0001"
>7_English/strings_s005.MSD.rpy- mix " “Kyaaa!”"
>7_English/strings_s005.MSD.rpy: voice "MIX0002"
>7_English/strings_s005.MSD.rpy- mix " “Kyaaaa!”"
>7_English/strings_s006.MSD.rpy: voice "MIX0003"
>7_English/strings_s006.MSD.rpy- mix " “Kyaaaa!♪”"
>7_English/strings_s007.MSD.rpy: voice "MIX0004"
>7_English/strings_s007.MSD.rpy- mix " “Kyaaaaaaaa!”"
>7_English/strings_s010.MSD.rpy: voice "MIX0005"
>7_English/strings_s010.MSD.rpy- mix " “ “Kyaaaaaaa!♪” ”"

Here, MIX0005 had a different character code. So I guess the first five should be labeled mix_ab and the sixth mix_ac.
I'm not sure if these quotes are right (they only happen on these Kyaa lines). I think they're partly also as parody, but two in a row can't be right.

SH9:

>9_English/strings_S005.MSD.rpy: voice "MIX0000"
>9_English/strings_S005.MSD.rpy- mix " “ “Kyaa~~~♪” ”"
>9_English/strings_S013.MSD.rpy: voice "MIX0000"
>9_English/strings_S013.MSD.rpy- mix " “Kyaaaa~~n♡”"

These two had different codes, so one should get labeled mix_ab and the other mix_ac.

SH10:

>10_English/strings_S007.MSD.rpy: voice "JMIXCD0000"
>10_English/strings_S007.MSD.rpy- girl_cd " “Kyaaaaaaaann♪”"
>10_English/strings_S009.MSD.rpy: voice "BMIXCD0000"
>10_English/strings_S009.MSD.rpy- club_member_cd "Whatttt!!"

These already have different labels, girl_cd (JMIX) and club_member_cd (BMIX).

>#TODO: There are also two other extra girl characters 女の子A (OnnaA) and 女の子B (OnnaB) that are distinct from the girls in the school. I believe they appear at s012. We will need a different label for them.

They have different labels already: woman_a and woman_b

>10_English/strings_S012.MSD.rpy- voice "ONNAA_0000"
>10_English/strings_S012.MSD.rpy: woman_a "Hey, hey, are you possibly a model?"
>10_English/strings_S012.MSD.rpy- voice "ONNAB_0000"
>10_English/strings_S012.MSD.rpy: woman_b "No way! You're not a model even though you're that cute? Since you have such a good fashion sense, you should definitely be scouted if you walk around the town like this!"
>10_English/strings_S012.MSD.rpy- voice "ONNAA_0001"
>10_English/strings_S012.MSD.rpy: woman_a "Your fashion sense is amazing~! Where do you buy your clothes?"
>10_English/strings_S012.MSD.rpy- voice "ONNAB_0001"
>10_English/strings_S012.MSD.rpy: woman_b "Hey, If you have time, you could come to tea with us. I have a lot of things I want to ask you."
>10_English/strings_S012.MSD.rpy- voice "ONNAA_0002"
>10_English/strings_S012.MSD.rpy: woman_a "What are you doing? …We're not finished talking yet."
>10_English/strings_S012.MSD.rpy- voice "ONNAB_0002"
>10_English/strings_S012.MSD.rpy: woman_b "That's right, don't interrupt!"
>> No.4476  

Hi, I see you still working on games or at least it looks like :P

I just want ask to make sure which games are fully redone for ren'py and can be translated without worrying of later changes in code.

>> No.4478  

>>4476
How much progress have you made in your translations?

>> No.4479  

Is there any way I can help with the projects so we can move foward and soon have the newest releases. I can offer Photoshop skills weather it involves cloning off text out of pages, adding text,logo making,or perfectioning breasts...if I knew what scripting is use or what is done I will so much be graceful to help. I Know css coding and html5 some, and flash coding.Is there a link to learn how to do this. Please send it to me and I'll teach myself. Thank you.

>> No.4480  

>>4479 I need to know the Japanese language right xD sorry for the bother.

>> No.4485  

>>4478
I didn't done anything more than 60 lines on beginning. I was waiting for you to finish rewriting because if changes would be too big in file structure I didn't want to copy and paste everything manually.

>> No.4487  

>>4485
The script is independent of the programming. You can continue translating as normal.

Ren'Py has a feature for multiple localizations which creates special files for alternate languages. However, it will be best to wait until a new translation and the program code are finalized. Until then, continue translating the UTF files as before.

>> No.4489  

I want to translate remake so I should download those files and replace whats inside "", right? https://github.com/AXYPB/pytals/tree/master/1_English_Remake

This "$ save_name = "After-School Shock"" is chapter name?

>> No.4490  

>>4489
These scripts contain only the primary language. Ren'Py can provide scripts containing localization commands to simplify translation. This is an example of such scripts for the Polish language for the first visual novel:
http://axypb.net/pub/SHP1_polish.zip

Insert your translations in the quotation mark pairs "" below each commented line of dialogue, as below:

# game/scripts/strings_S001.MSD.rpy:12
translate polish start_5c68a151:

# "It happened in April, as the sakura trees danced."
""

You may begin translating this if you wish, but changes made to the code will result in changes to these scripts, which will require manual copy and pasting of translated lines to the new versions.

>> No.4491  

>>4490

So what should I do to avoid possible manual copy lines?

>> No.4492  

>>4491
Continue working on the UTF files as if translating the original visual novel program.

>> No.4493  

One more question. Will it be okay if I put my translation in english translated UTF files or it has to be japanese because it won't be possible to covert them later for Ren'Py?

>> No.4494  

>>4493
You can work on the English source UTF files. If it becomes necessary, I will be able to copy your translation into the Japanese files.

>> No.4495  

That's great because earlier I had opened japanese and english file and it wasn't comfortable.

Thanks for everything and sorry for all those questions. I just didn't want to derp so badly that I would regret.

>> No.4566  

Just to let you know that I didn't resigned from translation I just translated 8th chapter. I will prabobly finish last 7 in a week but just for sake of my own peace of mind I'm counting to finish it in 2 weeks.

>> No.4568  

>>4566
Thank you. Please continue at your own pace.

To expedite the development of this program, I will focus on the first visual novel when I am able to resume. I apologize for the delay.

>> No.4591  

kostek00 has completed the first draft of the Polish translation and now requires it inserted into the Ren'Py working translation.

I have uploaded the translation and the Ren'Py localization template here: http://www.mediafire.com/download/3h01u30jkiycdpr/SH1-polish.zip

To create a localization, the quotation mark pairs in the Ren'Py scripts are filled with the corresponding line in the translation.

# game/scripts/strings_S001.MSD.rpy:12
translate polish start_5c68a151:

# "It happened in April, as the sakura trees danced."
""

It should be possible to automate this by using sed or AWK to read the contents of the UTF files and replace each "" in order. Such a command should read lines from a UTF file and for each "" in the RPY file, replace with the corresponding line, then repeat with the next line in the file.

However, the menus are not fully implemented yet and will need to be translated separately.

>> No.4592  

>>4591
I compiled the Polish rpy files:
http://www.mediafire.com/download/re51v4522mh4jee/Polish_SH1_v1.rar

I found one missing line in S003:
"You mean it? You'll call me Onee-sama?"

> # "I was weak against Yuuna-senpai even without her doing that."
> "Nawet bez tego nie potrafiłabym."
> # voice "YUNA0178"
> # yuuna "You mean it? You'll call me Onee-sama?"
> voice "YUNA0178"
> yuuna "「Ech...」"
>(Ech is the next line, so this translation is missing)
> # voice "NANA0210"
> # nanami "Yes..."
> voice "NANA0210"
> nanami "Przegrałam."
>(Przegrałam is "Utter defeat." (the next line)

The choices are also missing from S003. S015 is missing the last string "#TODO: Ending". Otherwise it looks good as far as I can tell.

The above missing translation I've replaced with "(missing translation: {english line})" so it isn't shifted like the above example.

>> No.4596  
File: 2015-02-13 01_31_00-A Kiss For The Pytals - autoreload.png -(655403 B, 808x627) Thumbnail displayed, click image for full size.
655403

>>4592
Thank you for your prompt response. I've instructed kostek00 to review your post to correct the missing lines.

>> No.4598  

I decompiled once again 3rd chapter and checked in game to make sure and remake version don't have that line.

>> No.4599  

Last line with Ending do exist it's last line in original script "<62D0>                      おしまい=                     Koniec".

>> No.4600  

One more thing I made mistake with name. I used Yuna instead Yuuna. I will correct it for final version.

>> No.4601  

>>4598
Oh right, I came across it before: >>4108
It should definitely be added though. The next line, Nanami's "Yes", is in response to this question.

>> No.4602  

I thought about it exactly the same because Nanami still gave her response.

>> No.4608  
File: strings_S003.MSD.rpy.txt -(40169 B, 0x0) Thumbnail displayed, click image for full size.

>>4601
The Polish s003 will need to be reprocessed with this correction.

>> No.4609  

>>4608
This is the same file as above.

>> No.4611  
File: Polish_S003.rpy.txt -(47686 B, 0x0) Thumbnail displayed, click image for full size.

Here is 3rd chapter with corrected missing line.

>> No.4620  

For those interested, I've built a standalone Ren'Py launcher with the Polish translation.
http://axypb.net/pub/SHP1-1.0-win.zip

You will need to provide the visual novel content from your own installation. To do this, use crass (http://tlwiki.org/index.php?title=File:Crass-0.4.14.0.bin.7z) to extract all files from the archives MGD, SE, BGM, and VOICE. Move all files out of the subfolders that crass creates into the game/common folder together.

Only the first branch has been implemented, and the menu screens have not been changed from the defaults yet.

>> No.4674  

kostek00 has completed the draft for the Polish translation and requires a new version of the Ren'Py translation files.
https://mega.co.nz/#!JxtHGQoR!ewfmCQJUEcUhfl2S93kODYXcjwPkQJ0gt0z9cJQ9K9A

S003 1 - missing line.txt contains the line "You mean it? You'll call me Onee-sama?" which was not present in the script provided.

>> No.4675  

>>4674
The files in the rpy folder from last time don't have an entry for chapter titles.

>> No.4677  

>>4675
Ren'Py's translation system does not handle the save_name variable at this time. I will ask the developer for a possible workaround.

>> No.4678  

>>4677
Okay, here's the updated strings files:
http://www.mediafire.com/download/27784z8bga4s0uo/Polish_SH1_v2.rar

>> No.4740  

i want translate Sono Hanabira ni Kuchizuke wo - Anata wo Suki na Shiawase. Please help me.

>> No.4741  

>>4740
I've uploaded a Ren'Py translation template for Anata wo Suki na Shiawase for an Indonesian translation.
http://axypb.net/pub/SHP5_indonesian.zip

Open the .RPY files in a text editor and insert your translations in the quotation mark pairs "" below each commented line of dialogue, as below:

# game/5_English/strings_S001.MSD.rpy:10
translate indonesian start_04d2b053:

# "It all happened so suddenly without the slightest warning."
""

If you require the original Japanese script, I can provide the .UTF files used for the original translation by request.

>> No.4742  

what should i do when doing compile again?

>> No.4743  

what should i do when doing compiling again?

>> No.4744  

>>4743
The files I uploaded are not for use with the original visual novels. I intended that you insert your translation into them while you run through the English translation of the visual novel in another window. This method will result in less manual work for you. You will not need to use ArcTool or do any compilation as we will no longer use them for the translations.

>> No.4745  

in that case, i c'ant take back the translate result to the game?

>> No.4746  

>>4745
No, your translation will be used in this project, which is built for the express purpose of simplifying translation work. When it is finished, you will be able to update the translation and correct errors easily without the need for compiling with each change.

>> No.4748  

correct if i'm wrong :
-this project is create new engine for sonohana
-this project is still develop

>> No.4750  

>>4748
You are correct on both points. The project is still active, despite a lack of public updates.

>> No.4752  

then what should i do if i finished the translation?

>> No.4753  

>>4752
Please gather your translation files into a ZIP and upload it to MediaFire.

>> No.5359  

Hello, Procyon. I'm the guy who made Polish translation for Sono Habanira. I know that project for moving this game series on Ren'py has been abondoned. I decided to continue it thou everything I'm rewriting to Ren'Py is made manually. If you are still here and would like to continue project your knowledge on pulling out strings from game as you did back then would really help me and save dozens if not houndreds of hours. Currently I rewritten something like 30% of first game.

If you decide to do that the best would be writing to at my mail "nakross@wir.pl". That way I will immediately know if you write something.

>> No.5360  

>>5359
What about the renpy code in the last English version I posted? I also made Polish files to go in the tl directory, but I could insert them in the renpy code if you prefer.

So you get something like this:

>label start:
> stop music fadeout 2.0
> scene BLACK
> with fade2
>
> $ save_name = "「Popołudniowy szok.」"
> stop music fadeout 1.0
> play music "common/M01.ogg"
> scene bg18a
> with GS_CUD
> window show
> "Wydarzyło się to w kwietniu, gdy zakwitały wiśnie."
> "Stałam w nowiutkim mundurku."
> "Nie byłam przyzwyczajona do niego, przez co ciągle zastanawiałam się, czy sukienka nie jest za krótka."
> "Odeszłam od reszty pierwszaków i zaczęłam ją poprawiać."
>
> show na tna03s with dissolve
> #GS_CLR
> voice "NANA0001"
> unknown "「Dlaczego mi się to przytrafiło?」"
>
> "Od dawna chciałam założyć ten mundurek..."
> "Ale nie przewidziałam tego."
>
> show na tna06s
> #GS_CLR
> voice "NANA0002"
> unknown "「No nie, wyglądam koszmarnie.」"
>> No.5361  

>>5360
If you could insert Polish into Ren'Py that would be great. The only thing I would need to do is to properly time those effects which I halfy done already.

I was also thinking to rename files and place them into different folders so for example speach would be in "voice" folder, backgrounds in "images/backgrounds", etc.

>> No.5362  

>>5361
All right, I have two versions of the first game, original and remake. Which should I use?

>> No.5363  

Remake, please. Let's stick to the newer version.

>> No.5364  

>>5363
↑ Forgot to put nickname

>> No.5365  

>>5364
Here you go:

http://www.mediafire.com/file/a76w19m95p2q3c7/Polish_tl_from_2015_03_25_rpy_from_2015_02_14.zip

>> No.5366  

This is amazing. I have everything that I was doing till now manually already in script. You don't even know how happy you made me. Thank you so much.

>> No.5367  

I'm wondering if you know what are the fonts names for title and it's subtitle in A Kiss For The Petals: Remembering How We Met?

>> No.5368  

>>5367
AXYPB says it's FOT-MatisseV Pro-EB for the main title and DFGPOPCorn-W12 for the subtitle.

>> No.5378  

I don't think I can get anywhere FOT-MatisseV Pro-EB. Even on sites that sells fonts and list this one don't offer it for sell.

>> No.5379  

>>5378
https://pan.baidu.com/s/1mi6eOHE
Is that it?

>> No.5380  

>>5379 From what I see it looks exactly the same. Thanks!

>> No.5382  

Procyon, most of game has been made (80%?). You helped me a lot that's why I would like to include you in credits for programming. Do you agree for that? I'm asking because not everyone might want that.

>> No.5383  

>>5382
Sure, I'm cool with that. Thanks.

>> No.5389  

Procyon, could you check file Polish_S010.rpy? First choice is missing 2 lines or last 2 from 2nd choice should be shown after any choice.

>> No.5390  

>>5389
I forgot what was up with the choices, but it looks like I wasn't able to extract the Japanese for them.

So the "Choice 1" and "Choice 2" in S003, S005, S008 and S010. Is there something special about S010? I don't understand what you mean.

>> No.5391  

>>5390

Currently is like this:

label b4b:
"Wazon czy doniczka z kwiatami miałyby szczęście, gdyby przetrwały tydzień."
show yuuna dressed 01 at cusleft
show nanami dressed 01 at cusright
#FadeHorizontalOut
voice "yuna0597"
yuuna "Poczekaj chwilę, przyniosę coś do picia."

voice "nana0655"
nanami "Dobrze."

label b4con:
#rightstayed
hide yuuna dressed 01
show nanami dressed 01 at cusright
with dissolve
#FadeHorizontalOut
play sound "sound/SE010.ogg"
#Music already playing: M02
"Zostałam sama w pokoju Siostry."

When checking original game it looks like last 2 lines from label b4b should be shown after both choices in label b4con.

Like this:

label b4b:
"Wazon czy doniczka z kwiatami miałyby szczęście, gdyby przetrwały tydzień."
label b4con:
show yuuna dressed 01 at cusleft
show nanami dressed 01 at cusright
#FadeHorizontalOut
voice "yuna0597"
yuuna "Poczekaj chwilę, przyniosę coś do picia."

voice "nana0655"
nanami "Dobrze."

#rightstayed
hide yuuna dressed 01
show nanami dressed 01 at cusright
with dissolve
#FadeHorizontalOut
play sound "sound/SE010.ogg"
#Music already playing: M02
"Zostałam sama w pokoju Siostry."
>> No.5392  

>>5391
You're right, reading the script it makes sense. I must have made an error in where the final branch label is put.

>> No.5394  

>>5392 Game is basicly ready to release. There is ability to archive it's files but I won't do that because I want to other individuals have easy way to translate game. But the problem will be language. As game currently stands it's main language is Polish and let's be honest I doubt anyone will know my language to translate from it. That's why I was thinking about making language selection and include english as well. Files with dialogs are checked, cleaned and prepared to match in 95% how everything is showed on screen with original game (around 5% will not match beacuse of engine differences). I was wondering if it will be possible for you to input english dialogs into script I prepared? If you could do that than after that I will need to generate tranlation files for Polish. There will be need to input polish dialogs into those files but if it's not possible to do that automaticly than I will have fun time with copy-pasting.

Translation files are different than normal files.
For example when normally this line looks like this:

> show nanami dressed 06
> with FadeHorizontalOut
> voice "nana0002.ogg"
> unknown "No nie, wyglądam koszmarnie."

In translation file it will look more or less like that:

> translate polish he85kc02:
> voice "nana0002.ogg"
> "No nie, wyglądam koszmarnie."

If it possible for you to do any of this above here is script I prepared: http://www.mediafire.com/file/b6z3ue0sf0oh5t0/SonoHana1_Polish.7z

>> No.5395  

I realized I will need only english dialogs inputed into script I provided above.

We won't need to copy-paste or input any dialogs into newly created translation file. Upon creating translation files it appears that Ren'py inputs original dialogs into translation files as comments. So after making translation files from Polish to Polish all I need to do is remove comments.

>> No.5397  

>>5394
>>5395
Do you want me to insert English lines into the above zip?

>> No.5398  

>>5395
Does this look okay?

http://www.mediafire.com/file/pcknryw693ct7w1/PL2EN.7z

(same as above, but without the Polish lines)
http://www.mediafire.com/file/2o62hoaa74wi30a/PL2EN_no_original.7z

Some lines that weren't in the English scripts are missing or still in Polish. (like choices, the "end" command and the game over message at the end of S012.

>> No.5399  

I can do the same for Japanese.

http://www.mediafire.com/file/g7s7kv1gs9667o2/PL2JP.7z

>> No.5400  

>>5398
Yes, yes, yes, it's awesome. This is exactly what I needed.

Missing lines are probably menu choices and one at the end of S012. Line in S012 is made by me and it's goal is to insinuate that it might not be true end yet. I will translate it later to English and Japanese (Japanese might be hard).

Choices in original game were made with images here they will be as text. It will be much easier to translate them.

>>5399
I'm not sure what to do with Japanese. On the one hand I don't want to compete with developer by including it as playable language. Competing isn't my goal. Also original version is playable on current OS's (well maybe except windows 10 not sure). On the other hand I would like to have as many languages as possible. I might also include Japanese but only for translation purpose. What do you think?

>> No.5401  

>>5400
It should be fine to do what you want until you get a complaint, I guess.

How are you sharing it when you're finished? I can link to it on the yuri-project.net homepage.

>> No.5402  
>It should be fine to do what you want until you get a complaint, I guess.

You are probably right.

>How are you sharing it when you're finished? I can link to it on the yuri-project.net homepage.

I'll post links on my website and they will be on mega and even maybe torrent (I linked my website in name) but it will be nice if you could like them as well. Before releasing game I will also make english for my website in case someome actually will be interested in transslating game to his language.

Do you want me to send links to you privately or should I just post them here when game will ba ready?

>> No.5403  

>>5402
All right, sounds good. Yeah, just post it here.

>> No.5404  

I don't know if those translations are fine but if anyone is willing to help checking them it would be awesome.

>From "Quick save complete."
>To "クイックセーブ完了"
>From "Failed to save screenshot as %s."
>To "スクリーンショットを保存できませんでした"

In sentence above I'm not translating whole only "Failed to save screenshot" part.

>From "Saved screenshot as %s."
>To "保存されたスクリーンショットを %s"
>From "Hmm... Is that really all?"
>To "うーん… 本当にすべてですか?"
>> No.5405  

>>5404

>From "Saved screenshot as %s."
>To "保存されたスクリーンショットを %s"

I think that should be
スクリーンショットは %s に保存された

>From "Hmm... Is that really all?"
>To "うーん… 本当にすべてですか?"

うーん…これは本当に終わりかな…?

>> No.5406  

>>5405

>From "Saved screenshot as %s."
>To "保存されたスクリーンショットを %s"
>I think that should be
>スクリーンショットは %s に保存された

With this one in that form it might be a problem beacuse %s shows not only files name but also path to it. With a lot of folders in path it can go beyond screen. Can this somehow be changed to have %s at the very end?

>: %s

Maybe like this?

>> No.5407  

>>5406
I think this should do
スクリーンショットの保存先は %s

>> No.5408  

Thanks a lot.

>> No.5409  

I need help with one more line. This is general message to show when there are missing other message. For example if there wouldn't be custom message for game saving or quitting game this one will be shown.

>Are you sure?
>> No.5410  

>>5409
本当によろしいですか?

>> No.5411  

>>5410 Thanks.

It seems like I can't just generate translations from Japanese and Polish and use them with English. I need to generate translations from English and then insert dialogs into those files for Japanese and Polish.

Here is example how header looks like:

>translate Polish start_5c68a151:
>translate Japanese start_5c68a151:

As you can see for both of them strings after "start_" are the same. It looks like Ren'py is genereting those string depending on what dialogs were in original language. So if I generate translation from Polish or Japanese those strings will be different.

Will it be possible for you to somehow insert dialogs into those translation files? At the end of those files are also strings translations. They are from menus and chapter's names. You can ignore them I can manually do them.
http://www.mediafire.com/file/phn4g3308c8gqe4/translations.7z

>> No.5412  

I somehow did it in a clunky way but I did it. Let's hope I didn't messed anything.

>> No.5418  

It is done. I finished game. Here are links:
Mega: https://mega.nz/#F!BeIw2aTA!gghh2hbro34Sg-mPMvQcTg
Torrent: https://sukebei.nyaa.si/view/2509346
(I have slow upload)

If it's not a problem could you also attach message from me? It's a little instruction in case someone will be interested in translating game on his language.

>"If you would like to translate this game on your language you are free to use anything you want from this release. But I recommend you contacting with me first. You can do this by discord (link to my channel is on the bottom of my website - preferred method) or by mail (mail is in FAQ on my website - I really don’t like writing by mail). I can help you with anything you will need and provide you psd files with all graphics that needs changing so you won’t need to do everything from scrach. Also it would be really awesome if all translations could be in one release rather than every in different one. Game is made to support multi-languages."

This is my website: http://nyan.skanlacje.pl

>> No.5419  

I forgot to remake torrent after changes, this one is good: https://sukebei.nyaa.si/view/2509352

>> No.5436  

Procyon, I was thinking about translating other games form this series. I mean those that already are fully translated to english. Would like to take part in that?

>> No.5437  

>>5436
Yeah sure. If you need anything, let me know.

>> No.5438  

>>5437
I have done few things I wanted and now I can work on another game.

I'll go with their release date. So next one will be Sono Hanabira ni Kuchizuke wo: Watashi no Ouji-sama. Same as first Sono Hana it was in renewal edition so I will use remake. Before you start your magic could you check if remake have any missing lines?

Don't remove any special characters like stars as it was in first Sono Hana. If they are there to show different character I'll change them manually. It's not a problem.

Also if you can don't replace origial names for transitions like "GS_HI_NS" with dissolve or any other effect. Only add "with" before them. I will create all those effects like in first Sono Hana (or atleast I'll try).

Please generate English and Japanese version of script. Same as in first Sono Hana English will be main language but I will need Japanese in order to pull out dialogs only in Ren'Py and then insert them in translation.

>> No.5439  

>>5438
The files I have from years ago are here:
http://www.mediafire.com/file/gjfkjcjk9djyakt/Sonohana2_english_rpy.7z

Can you give examples for how that should be adjusted?

I don't have Japanese MSD files besides Sonohana 1 remake and 11.

>> No.5440  

>>5439

>show sa tsa02s at center
>with dissolve
>#GS_CLR
>voice "JOSC0001"
>girl_c "Her face is small like a doll's~"
>show sa tsa01s
>#GS_CLR
>voice "SARA0002"
>unknown "Yes, I got a day off today."

Take a look at this part for example. GS_CLR is commented out or replaced by "with dissolve". If you can could change those to only contain "with GS_CLR" without commenting it out? GS_CLR is effect where something is disappearing from middle to sides horizontally. From what I saw only this one effect is changed that way.

Aside from this one everything else looks good.

>I don't have Japanese MSD files besides Sonohana 1 remake and 11.

That's not a problem. I will pull them from game and prepare to insert into translation.

>> No.5445  

>>5440
I'll look into it soon.

>> No.5448  

>>5440
Which one of these three ways should I do lines where dissolve is after a keyword:

>strings_S015.MSD.rpy: show sa tsa08s #with dissolve
>strings_S015.MSD.rpy- with GS_CLR
>strings_S015.MSD.rpy- voice "SARA1071"
>strings_S015.MSD.rpy- sara "Hmm."

or

>strings_S015.MSD.rpy: show sa tsa08s
>strings_S015.MSD.rpy- with GS_CLR
>strings_S015.MSD.rpy- voice "SARA1071"
>strings_S015.MSD.rpy- sara "Hmm."

or

>strings_S015.MSD.rpy: show sa tsa08s with GS_CLR
>strings_S015.MSD.rpy- voice "SARA1071"
>strings_S015.MSD.rpy- sara "Hmm."

>>5438

>Don't remove any special characters like stars as it was in first Sono Hana.

They seem to be removed in the above archive, right? Can you give an example line where one is missing?

>> No.5449  

>>5448

Please do second option.

>>strings_S015.MSD.rpy: show sa tsa08s
>>strings_S015.MSD.rpy- with GS_CLR
>>strings_S015.MSD.rpy- voice "SARA1071"
>>strings_S015.MSD.rpy- sara "Hmm."

When "show" command is in the same lines as "with" (show sa tsa08s with GS_CLR) then effect is only applied to this specific image (tsa08s). But when "with" is in new line then effect is applied to everything that requires changing, scenes, characters, etc.

>>Don't remove any special characters like stars as it was in first Sono Hana.
>>They seem to be removed in the above archive, right? Can you give an example line where one is missing?

I checked it to be sure and they aren't removed at least not stars and notes. I don't know if there are any other special characters. I assumed they will be because in first game they were removed.

>> No.5450  

>>5449
http://www.mediafire.com/file/d177f754fmb7tl7/Sonohana_02_Adjustments.7z
Let me know if you need anything else.

>> No.5462  

I found bug in script. "—" are replaced by question marks or double dashes. You don't have to remake script I will fix them manually while I'm checking script anyway. It's just info for future string pulling.

>> No.5533  

Hello, Procyon.
I released 2nd game and I would like to go with 3rd one. Are you up to another one?

>> No.5534  

>>5533
Nice, well done.

I've made the same adjustments to 3: https://www.mediafire.com/file/ty1ws53dwf6zyz7/Sonohana_03_Adjustments.7z/file

If anything is wrong, let me know.

>>5462
I changed the double dash to a long dash. If more special characters are wrong I'll need an example line.

>> No.5537  

Can I ask you to modify your script with following changes? It's fine if you can't change them.

  1. Files options and screens aren't needed. I have my own with everything that's needed. Defines file will be useful.
  2. Add extension to voice tracks so they will look like that "voice "MAI0002.ogg"". I have in mind something special when I release more games from series and I used voice playing without option that automaticly searches for ogg file with that name and thus it won't be compatible.
  3. Change names for placing characters on sides of screen from "l" to "left" and from "r" to "right".
  4. Change names of transistions effects. It's easier for me to manage them by those names rather than original names. Files names are still the same it's just name in code. Here is list:
> FadeHorizontalOut from GS_CLR
> FadeVerticalOut from GS_CUD
> FadeVerticalIn from GS_UDC
> FadeRingOut from GS_EN_NS
> FadeRingIn from GS_EN_SN
> FadeClock from GS_R
> FadeSquareOut from GS_HI_NS
> FadeLeft from GS_RL
> FadeRight from GS_LR
> FadeUp from GS_DU
> FadeDown from GS_UD
> Fade2 from fade2 (unfortunately it does matter)

5.

>I changed the double dash to a long dash. If more special characters are wrong I'll need an example line.

Can you also change all short dashes "-" to long dashes "–"?

6. You can remove path to files like for example "common/file_name.ogg" I set up automatic search for paths.

7. When showing character use full name for them. Example: "show re tre03s" to "show reo tre03s". "show ma tma08s" to "show mai tma08s". Same change should be made in define file.

8. When there is scene change or showing image named as "cu<number>" (they are chibi usual) they should be preceded by "window hide" command and ended with "window show" command. So if you have something like that for example:

> scene BG20a
> show cu04
> with GS_R
> voice "REO0054"
> reo "I'm not a first-yeaaarrrrr!!"

It should be changed to:

> window hide
> scene BG20a
> show cu04
> with GS_R
> window show
> voice "REO0054"
> reo "I'm not a first-yeaaarrrrr!!"

Whole code block should be included into them but should be ended before first text block or voice command if line is dialog.
When it comes to scene changes it might be tricky because code above should be applied only when new scene is showed not when scene command is used. So when scene is used but with the same image as previous it shouldn't contain window hide/show command. I understand that this is probably very hard to make in script so don't hesitate not doing it.

9. Is it possible for you to track where reaply starts and ends and write "#replay start/end"? They are accessed from extra menu in memories section. If not I will search for them manually.

Remember, if you can't do those changes it's fine.

>> No.5539  

>>5537
https://www.mediafire.com/file/t1xaogn62c9z881/Sonohana_03_Adjustments.7z/file

I gave it only a cursory glance after running the script, so if I missed anything let me know.

Note that some minus signs in defines.rpy got replaced with long dashes.

>When there is scene change or showing image named as "cu<number>" (they are chibi usual) they should be preceded by "window hide" command and ended with "window show" command.

I tried to do this using anchors of "scene BG" and a voice line or narration. If specific cases are wrong, it's easier if you adjusted them manually.

>Is it possible for you to track where reaply starts and ends and write "#replay start/end"?

No, I don't think I found the binary codes for those.

>> No.5540  

>>5539
Can you reupload file? It throws error that it's not available.

>I gave it only a cursory glance after running the script, so if I missed anything let me know.
>Note that some minus signs in defines.rpy got replaced with long dashes.

A will keep that in mind.

>>When there is scene change or showing image named as "cu<number>" (they are chibi usual) they should be preceded by "window hide" command and ended with "window show" command.
>I tried to do this using anchors of "scene BG" and a voice line or narration. If specific cases are wrong, it's easier if you adjusted them manually.

Not a problem.

Thanks for all your work.

>> No.5541  

>>5540
https://mega.nz/#!WVwA0Y7b!lnfgomDXT9gCjHnnDQi8GA3kv5jTact9wdoF1iPYGXU

>> No.5542  

>>5541
Thanks.

And I forgot about one more thing.

>7. When showing character use full name for them. Example: "show re tre03s" to "show reo tre03s". "show ma tma08s" to "show mai tma08s". Same change should be made in define file.

The same should be applied when hidding character. You don't have to make another script I will make that myself.

>> No.5543  

>>5542
I can change that really quickly.

https://mega.nz/#!nRAT3YDT!VAZF_PX2F6jBXbE7NXx_kAmsXJWKKoWTSA7ncyXLofI

>> No.5546  

>>5543
Thanks.

>> No.5602  

Hello, Procyon. Are still on this forum?

It's been a while but I finally finished and released 3rd game. Soon I would like to start working on next game from series Sono Hanabira ni Kuchizuke o: Itoshisa no Photograph. Would you like to help me with another game?

>> No.5603  

>>5602
Sure, then I'll just run the previous script on game 4:

https://mega.nz/#!jEBGRK5K!530nnwgSC7K3JX270veW_aF8ud-W3e7WSCcz-3Xv1o0

Let me know if you need me to change anything.

>> No.5604  

Alright, thanks.

>> No.5609  

In case anyone is interested in rewriting project. I didn't abandon it. I just didn't had time for it but I finally moved with it.

I'm not going to go in details what I'm changing because you will be able to see changes after rewrite is released. (If you know any coding languages you will see how much is changed.) But for 4th game I'm doing major code rewrite. Future releases will also use this code and files managing. This will be beneficial for potential translations for other languages and easier gui changes.

I'm expecting to finish rewrite with translation to my language in February.

>> No.5619  

Not quite February but I finally finished game. Here are links:
Mega: https://mega.nz/#F!Wq5HkYYY!ASaRNjkoH2nlGeaXKMcFjA
Torrent: https://sukebei.nyaa.si/view/2972027

>> No.5621  

Great work, everyone.

>> No.5622  

>>5619
Good job.

>> No.5623  

>>5622

Thanks and you too. Without you it would take so much longer.

>> No.5627  

Hello, Procyon. I would like to soon start rewriting next game in series. It will be "Sono Hanabira ni Kuchizuke o: Anata o Suki na Shiawase". Could you prepare script?

>> No.5628  

>>5627
Yeah, no problem:

https://mega.nz/#!SEJ2kQTQ!EKB0aKX2LDUiDkEwtLcN5T-fbnDuUrPIowYhWi01aGE

>> No.5629  

I would have done it sooner if I had noticed your post. Just in case, I'll do the others right now as well.

6 adjustments: https://mega.nz/#!2AJG0CYD!QmdPTrQTsBl3a7eWQ9kKup6Daivi81yoKllycoU3mKM

7 adjustments: https://mega.nz/#!bYYwFC5D!0WS9ObXCNLCOvwXZV7APChyL3RlH25U3ED4p67dRhmw

8 adjustments: https://mega.nz/#!bAQE2IKT!WuRWW9JrNssmZY3Zu6y-7aXbLj6rAfn40L8WlQJsXqY

9 adjustments: https://mega.nz/#!TVIGVCgT!Bub-_Ke6mz18cjoCl5rR24bLfx5iib_oMr3DR_6StfU

10 adjustments: https://mega.nz/#!aNpxyLRZ!Z71dTXSjB0xpz1R8fJzmnpnEu-0I9QeeL6Uiciry9Z8

>> No.5630  

Thank you.

>> No.5631  

Could you check code for choices in Kuchibiru to Kiss de Tsubuyaite (SH06)? In your adjustmens it says player has to choose 1-2-1-2 answers to pass. However upon playing original version it seems like it is 1-2-1-any. I tried many variants of choices and it only allowed me to pass beyond chapter 12 if I didn't made any mistake in answers 1-3.

>> No.5632  

>>5631
I wasn't able to extract the logic for the score directly from the msd, so I used a walkthrough and it says 1-2-1-2, but I suppose it must have just recorded one possible route.

>> No.5633  

Oh I see. I have to play through them anyway to correct any bugs and add new crap they come up with so I will just check this as well while I'm at this.

>> No.5634  

>>5632

For what it's worth, a redditor has posted the choices and number of incorrect answers allowed for every VN at https://www.reddit.com/r/SonoHanabira/comments/bt22gu/sono_hanabira_definitive_full_series_walkthrough/

>> No.5635  

I saw it already but they are not 100% correct.

>> No.5636  

I can't believe how many bugs there is in SH07. They had working code from previous games and yet they fucked up so horribly.

Anyway... Procyon, would it be possible for you to add "with FadeHorizontalOut" after every group of showing characters? Examples:
Change:

> show runa

Into:

> show runa
> with FadeHorizontalOut

Another example:

> show runa
> show takako

Into:

> show runa
> show takako
> with FadeHorizontalOut

While original game doesn't have characters transition effect from what I checked this is the only game that doesn't have it. I will assume they let some collage student do this and he was lazy enough to not add them.

>> No.5637  

>>5636
Yeah, I can do that. Just to SH07?

>> No.5638  

>>5636
https://mega.nz/file/PMpEhKYK#elbvx2HtkZqzMrjT-aJjSR81b43zKn8EDpRmvqM9p8E

>> No.5639  

Thanks.

>> No.5642  

I'm wondering what is the logic used in Sono Hana games. I'm now checking in SH07 what are correct answers. Checked all combinations and got this:

>> 1111 x
>> 1112 x
>> 1121 v
>> 1122 x
>> 1211 x
>> 1212 x
>> 1221 x
>> 1222 x
>> 2111 v
>> 2112 x
>> 2121 v
>> 2122 v
>> 2211 x
>> 2212 x
>> 2221 v
>> 2222 x

Combinations with "x" are incorrect, combinations with "v" are correct. If those games uses point system some of those answers shouldn't work.

>> No.5645  

>>5642
In this case every correct answer is worth 1 point, with 2121 being the correct answer.
3 points or above are required, so combinations 2122, 2111, 2221 and 1121 will also work.

>> No.5646  

This would be true if both answers are correct which they aren't. I suspect they hard coded correct combinations instead using point system.

>> No.5649  

>>5642
hello. i have heavily researched the first game's engine, which the seventh game's appears to be extremely similar to.
in short, anon above is correct and he does not contradict your findings.

now, for the main point of this post.
here is an exclusive, never before seen, readable version of the seventh game's scripts. follow along for maximum fun.
https://mega.nz/file/AkkS2ADI#KdEWko_6Fflc79ArNYqKCTdpwrBLKHc1pWEsD-uMuns
open the files in a text editor.
because this is an unofficial script decompilation, all the command names are based on research.
look in the __commandNames file for a list of all known commands. in the scripts, any command that is a number instead of a name is an unknown command.

to illustrate the choice system, we'll look at the second choice which is in s006.
first find the "showChoice" command. then look at the "jumpList" command under it.
here, the jumpList command goes to Label 1 if you chose the first choice and Label 2 if you select the second choice.
those label numbers are determined by the 1 and 2 you see on the "jumpList" line.
to see the code for the first option, search s006 for a line that says "label 1".
if you follow the code down from there, you eventually find code that says:

>setVar 150 ((vars[150]) + (1))

that line increments variable #150. this variable stores the point count.
follow farther and you'll find "jumpLabel 3". that code jumps to the common point.
to see the code for the other option, search for a line that says "label 2".
if you follow down, variable #150 is never incremented and the code eventually reaches the common point without a jump. this is why the second option is wrong.
the correct choices are, as anon said above, 2121.

next, go to s015. we have a line near the bottom that says:

>ifStatement vars[150] ">=" 3 2

this means:

>if variable #150 is >= 3, jump to label 2.

because variable #150 is the points, this checks if you have at least 3 points.
if you follow label 2, it eventually jumps back to the main flow script. this continues the game.
if the if statement condition fails, no jump is performed. continue down and "returnToTitle" is eventually called.

for bonus fun, look at main.msd. there's debug code to display the current value of variable #150.
according to google translate, it's called "favorability"

>> No.5650  

I have no idea why I thought Anons answer was incorect. Now that I'm looking at my own answer everything is correct. Game uses the same point system as I did for renpy. Sorry, Anon.

>>5649
Do you think you woukd be able to decompile files for all Sono Hanabira games that were using MScript engine? I would love to code in exactlty which answer is correct because as of now I'm doing it by tries and errors method.

>> No.5651  

>>5650
yes. i will do it in a few days.

>> No.5652  

>>5651
Thanks.

>> No.5653  

>>5652
here you are, this should be all of them. good luck.
https://mega.nz/file/89EBhaAI#9nA-80nQlExk3Vb4QWsr9eV3agP17phZISpLcwvf6M0

>> No.5654  

>>5653
Thank you very much.

>> No.5656  

>>5653

I saw you also managed to decompile remakes for 1-3. I just checked my files and it looks like game 1-2 remake version have disappeared from my HDD. Would you perhaps be able to send me those 2 games?

After finishing 10th game I will most likely update games 1-4 with current game code and I might need those 2 games because for games 1-3 I used remake versions.

>> No.5657  

Thanks to user "aaaaa" I gathered all actually correct choices for games 1-10 not just guess. They are 100% correct with walkthrough from this reddit https://www.reddit.com/r/SonoHanabira/comments/bt22gu/sono_hanabira_definitive_full_series_walkthrough/

SH01: 2-1-1-1
Minimum points: 3
Allowed mistakes: 1

SH02: 1-1-1
Minimum points: 3
Allowed mistakes: 0

SH03: 1-1-1-1
Minimum points: 3
Allowed mistakes: 1

SH04: 1-any-1-1-1
Minimum points: 3
Allowed mistakes: 1

SH05: 1-1-1
Minimum points: 3
Allowed mistakes: 0

SH06: 1-2-1-1
Minimum points: 3
Allowed mistakes: 1

SH07: 2-1-2-1
Minimum points: 3
Allowed mistakes: 1

SH08: 1-1-2-1-1
Minimum points: 4
Allowed mistakes: 1

SH09: 1-2-2-1-1
Minimum points: 3
Allowed mistakes: 2

SH10: 1-2-1-2
Minimum points: 3
Allowed mistakes: 1

In case someone will need those decompiled files 01-remake and 02-remake are by mistake swapped.

>> No.5658  

>>5656
sure.
https://mega.nz/file/i6wFySAb#umhLB-A9yuUeWj9PZ-S9JTFbruV1HMo3m8CYx8BkuTg

>> No.5659  

>>5658
Thank you so much. You're saving my ass.



Delete Post []
Password