lsm/mdbtools
1
0
Fork 0
mirror of https://github.com/mdbtools/mdbtools.git synced 2026-01-21 18:48:34 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Better compressed text handling

Browse Source
This commit is contained in:
whydoubt
2004-12-11 06:07:20 +00:00
parent fa8d24dd2b
commit d271b5fae5
10 changed files with 192 additions and 89 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
HACKING
View File

@@ -697,3 +697,27 @@ Next comes one of more rows of data. (column names, descriptions, etc...)
See kkd.c for an example, although it needs cleanup.
Text Data Type
--------------
In Jet3, the encoding of text depends on the machine on which it was created.
So for databases created on U.S. English systems, it can be expected that text
is encoded in CP1252. This is the default used by mdbtools. If you know that
another encoding has been used, you can over-ride the default by setting the
environment variable MDB_JET3_CHARSET. To find out what encodings will work on
your system, run 'iconv -l'.
In Jet4, the encoding can be either little-endian UCS-2, or a special
compressed form of it. This compressed format begins with 0xff 0xfe.
The string then starts in compressed mode, where characters with 0x00 for the
most-significant byte do not encode it. In the compressed format, a 0x00 byte
signals a change from compressed mode to uncompressed mode, or from
uncompressed mode back to compressed mode. The string may end in either mode.
Note that a string containing any character 0x##00 (UCS-2) will not be
compressed. Also, the string will only be compressed if it really does make
the string shorter as compared to uncompressed UCS-2.
Programs that use mdbtools libraries will receive strings encoded in UTF-8 by
default. This default can by over-ridden by setting the environment variable
MDB_ICONV to the desired encoding.