НОВОЕ: OS/2 GURU - Вопросы и ответы

Пакет русификации для OS/2

Operating systems:
ArcaOS, eComStation, IBM OS/2 Warp
eComStation myths 

State of LIP  





LIP FAQ (Internal)

Q1: Algorithm of LIP processing

Current technology:
1. Install MCP of language LANG to Virtual Machine
2. Collect all files and object names from this copy
3. Build the list of object names 
4. Compare all files of MCP/LANG  and english files  byte by byte. 
      Similar files are removed.
5. If files are different:
      compare versions, check translated resources, extraction and copy 
      of resources. collect processed files to archive.
6. The files which are missing in eCS but are available in MCP - think, 
      is it necessary keep this file?
      analyze this files. Collect this files to other archive.
7. Processing of config files, extraction of texts. 
8. Compilation of LIP package

* if several LIP packs are processed, the next package requires less time.

Q: What tools, scripts does eCo Software use?

A: We have used this tools in the past: (rus) http://ru.ecomstation.ru/showarticle.php?id=197

Q2: Why install OS to collect translated files?

A2: It's difficult extract data from packed files, from installers. It's easy collect data from working OS.

// Of course, it's necessary install all components

links: 20120217

Q3: What scripts are used to collect data for LIP?

A3: Now, when we have several LIP packages (russian, german, portuguese) this can be automated again.

! the experience with german (after russian) demonstrates that the scripts don't solve all problems, every language requires additional manual efforts.

links: 20120217

Q4: How many text files, binary files, total Qty of files?

A4: Example: russian LIP

Total quantity of files in LIP package: 1230

* text files: ~525 (*.txt, *.doc, *.cmd, *.htm, *.nif, *.rc and other,
   including *.msg, *.inf, *.hlp, *.lng and other, which require compilation

* binary files: ~540 (*.dll, *.exe and other);
* files which don't require translation: ~165  (fonts, vocabularies, syslevel.* 
and other files which are not translated).

So.. text/binary files: 50/50.

links: 20120308

Q: What scripts do you use?

=== We have this scripts: ===

1. Scripts to process names of WPS objects:
* collect info from all files folder.obj in sub-directories to one file
* collect OBJECTID and name from WPS backup (3, for different cases)
* substitute translated names by OBJECTID

2. Several scripts to process files:
* Check presence of resources in the files in current sub-directory and all sub-directories, files without resources are removed.
* Extract resources from files in current directory and all sub-directories
* Script to compile resources (don't spend time to switches and pathes)
* Set date and time for files in the directory and sub-directory by sample

=== What could be automated ===

We think that all routine operations are automated.

We are missing script aimed to compare RC/DLG/MSG/NIF files with consideration of different languages
and possibility to edit information.




(C) OS2.GURU 2001-2021