| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
Note that there's not a space before the (2): it seems that strings
like that are used for Copyright (C) etc. and not links).
|
| |
|
|
| |
Note: the stray g_next_char() line had no effect.
|
| |
|
|
|
|
|
|
| |
Since we're parsing the whole thing in one pass, we can't know this is
the last line until we're done, so we go through at the end and check
whether there's an empty section. If so, we know what the title should
look like, so we extract the information and put it somewhere more
useful.
|
| | |
|
| |
|
|
|
|
| |
This works harder to understand the first line of output: the previous
code didn't work if there was a space (or hyphen or anything else) in
the command name.
|
| |
|
|
|
| |
This allows us to work out when there should be a wider gap between
two paragraphs, for instance.
|
| | |
|
| |
|
|
|
|
|
| |
At the moment, there's a horrible hack with a hardcoded table of
character names and unicode code points. Hopefully eventually this can
be replaced by parsing a file or calling a program... but I don't know
how yet :-(
|
| |
|
|
|
| |
Basically, to get internal spaces correct (eg in tables etc), we
*have* to use a monospace font.
|
| |
|
|
|
|
|
|
| |
Currently doesn't deal with 'special characters' (including umlauts
and ... hyphens!)
Also seems to fail with pretty simple formatting (see list of sections
in man man)
|
| |
|
|
|
| |
Since GDataInputStream doesn't return newlines, we end up running words together
if the whitespace between them is a newline. This approach might be overzealous.
|
| |
|
|
|
| |
Lines from GIOChannel do. A bunch of the code from the old GIOChannel-based
parser was wrong, because it was looking for \n.
|
| |
|
