How to Create a DSL dictionary for Goldendict

[ikm]

You can try StarDict file format. It supports synonyms like in bgl, and also arbitrary html, too. The format itself is binary, but documented, and It has a "babylon" tool which converts .gls to stardict.

[panho10]

The reason I prefer dsl is that it can be used as uncompiled plain text. A large data takes a long time to compile, and further I can change or add articles anytime with no need for compiling it. Therefore if the data is not fixed and piled up as you proceed with your work or study, dsl or gls is very convenient.

[jennie]

I just wanted to make certain about this: Does goldendict support DSL dictionaries in UTF8 encoding or is it only UTF16? UTF8 doesn't seem to work for me, but I'm wondering if I 'm doing something wrong.

[ikm]

It is supported, but you need to begin file with UTF-8 BOM (EF BB BF).

[jennie]

Thanx a lot! This worked great.

[ikm]

You are welcome.

[francesinha]

It is supported, but you need to begin file with UTF-8 BOM (EF BB BF).

Hello ! I'm sorry, I don't understand this.

If I go to http://www.dicts.info/uddl.php, get the Wikipedia interlanguage links between English and French, for example, paste that content in a text editor, add the line breaks and the tabs, save the file as a .DSL...

1) Do the line breaks need to be Windows-like ? (I'm running XP, but my text editor says the file is saved with UNIX line breaks, not sure why)

2) Does the encoding have to be something in particular ? (my default seems to be "Windows 1252")

3) Is this what should be at the beginning of the file ?

#UTF-8 BOM (EF BB BF)
#NAME "Wikipedia interlanguage links English-French"
#INDEX_LANGUAGE "English"
#CONTENTS_LANGUAGE "French"

Thanks !

[jennie]

1) Do the line breaks need to be Windows-like ? (I'm running XP, but my text editor says the file is saved with UNIX line breaks, not sure why)

2) Does the encoding have to be something in particular ? (my default seems to be "Windows 1252")

3) Is this what should be at the beginning of the file ?

1) No, line breaks can be anything.
2)You must choose encoding: UTF-8. If you are using Windows, I think the BOM gets added automatically. So, your dictionary should work with goldendict. If it doesn't, well, I'm not a regular windows user, so I don't know any more professional ways to do this, but here are some suggestions:
The easy way (which will take more space in your hard disk) is to use UTF-16 encoding, and do nothing else.
One free text editor that I have used (geany includes an option to add the unicode BOM (Document->Write/Add unicode BOM). Maybe your editor has something like that, or maybe you'd like to download geany...
3)No, it should look normal, i.e.:

Code: Select all

#NAME "Wikipedia interlanguage links English-French"
#INDEX_LANGUAGE "English"
#CONTENTS_LANGUAGE "French"

The BOM is invisible. Like a paragraph mark, it doesn't take any space in your document.

[francesinha]

Thanks a lot, Jennie ! It works !

[francesinha]

Me again with the n00b questions... Well, first, allow me to wish you all a Happy New Year!

And now for the questions:

1) how do you insert a line break in a definition ? Is it possible to insert a line break in a headword ?

There seems to be a limit to the length of headwords and definitions.
2) Is it a .DSL limit or a Goldendict limit ?
3) What is it : how many characters/words ?
4) Is it possible to override it ?

Thanks !