Commit graph

29 commits

Author SHA1 Message Date
Eric House
35bdd2282e limit pattern parse by max length of tiles 2020-12-27 11:53:36 -08:00
Eric House
d0f5d9949a handle two-letter tiles better
Choose the two-letter tile where possible. Not sure though if this is
always the right move.
2020-12-26 20:45:34 -08:00
Eric House
2b56b1f8b2 cleanup 2020-09-29 16:06:59 -07:00
Eric House
eb728e4368 fix crash: don't go beyond last edge
An obscure pattern from a user caught this one, but the test should have
been there all along.
2020-09-28 19:18:09 -07:00
Eric House
e9ec30206d add debug-only field tracking place in iterator 2020-09-28 19:18:04 -07:00
Eric House
8511c633ee Merge android_translate branch 2020-08-12 16:50:34 -07:00
Eric House
7a43e95aa8 fix crashes when filter strings are too long 2020-08-11 14:13:11 -07:00
Eric House
b8f359c3e5 add filtering to wordlist browser
Add a basic regular expression engine to the dictiter, and to the UI add
the ability to filter for "starts with", "contains" and "ends with",
which translate into ANDed RE_*, _*RE_* and _*RE, respectively (with
_ standing for blank/wildcard). The engine's tightly integrated with the
next/prevWord() functions for greatest possible speed, but unless
there's no pattern does slow things down a bit (especially when "ENDS
WITH" is used.) The full engine is not exposed (users can't provide raw
REs), and while the parser will accept nesting (e.g. ([AB]_*[CD]){2,5}
to mean words from 2-5 tiles long starting with A or B and ending with C
or D) the engine can't handle it. Which is why filtering for word length
is handled separately from REs (but also tightly integrated.)

Users can enter strings that don't map to tiles. They now get an
error. It made sense for the error alert to have a "Show tiles"
button, so there's now a dialog listing all the tiles in a wordlist,
something the browser has needed all along.
2020-08-05 09:47:44 -07:00
Eric House
0e9661aa19 fix search of wordlists containing duplicates
Hungarian is unique (so far) in having two-letter tiles that can be
spelled with one-letter tiles AND in allowing words to be spelled both
ways. This crashed search based on strings because there were
duplicates. So now search is done by tile arrays. Strings are first
converted, and then IFF there is more than one tile array result AND the
wordlist has the new flag indicating that duplicates are possible, THEN
the user is asked to choose among the possible tile spellings of the
search string.
2020-05-04 08:33:15 -07:00
Eric House
98ce0e416f api cleanup: dict_iter -> di
(JNI changed too, so it's not just compiler-checked, but tests ok)
2020-05-03 21:45:08 -07:00
Eric House
353e9cf795 remove assertions from dictiter code on Android
Makes wordlist scrolling really slow, and since it's common code keeping
the assertions in Linux is probably good enough.
2020-05-03 10:18:37 -07:00
Eric House
e34da24393 revert wordlist changes
There are bugs there still to resolve and I need to ship. Will return
once they're fixed.
2020-04-29 10:15:58 -07:00
Eric House
8097a6ba0b fix not indicating when no results 2020-04-28 15:24:32 -07:00
Eric House
e7c96290ea name change only
It was confusing using dict_ as the prefix for two "classes"
2020-04-28 15:24:22 -07:00
Eric House
b2bc4e34cf do wordlist lookups using tiles not chars
Fixing a problem with languages (like Hungarian) where it's legal to use
a two-letter tile or two single-letter tiles to play the same word. When
words are seen, or searched for, as char-arrays, there are
duplicates. Current code crashes, but there's also risk the user gets
unexpected behavior. Crash is fixed, and foundation laid for better UX,
by doing all searches for tile arrays. If a search string translates to
more than one tile array the user must choose. For that choice to make
sense it's now possible to translate tile[] to char[] with a delimiter
between the tile strings.
2020-04-28 13:29:12 -07:00
Eric House
3a628f4375 load new-format dict into linux client, display default faces
correctly, and search using alternate as well as default faces.  Next:
support for alternate specials.
2013-04-06 11:43:57 -07:00
Eric House
4cb848582e fix edge cases in prefix matching 2012-08-28 07:35:11 -07:00
Eric House
145317426d Fix prefix matching to accept letters constituting the first part of a
digraph tile when at the end of a prefix so that e.g. GORIL in Catalan
will list GORIL·LA (rather than nothing since GORIL, ending with the L
tile, is not a prefix.)
2012-08-27 21:50:09 -07:00
Andy2
d856b9a8ce Don't behave badly in presence of empty dict: word count should be 0
and sanity test should pass.
2011-11-21 18:06:46 -08:00
Andy2
cc0105e14c add param to dict_countWords in which it accumulates counts per word
length.  Shouldn't hurt speed too much if null passed.
2011-11-18 08:15:21 -08:00
Andy2
598be8b404 MAX_COLS -> MAX_COLS_DICT for dict-related stuff (since dicts are
staying with fewer words for now.)
2011-11-18 07:56:02 -08:00
Andy2
3707e7aaca tweak to get performance to same level as before length-filtering was
added, mostly by using local variables instead of ptr-accessed in
struct and macro magic to support that.  Passes every test I can think
of on Linux but fast-scrolling broken on Android.
2011-11-05 06:14:35 -07:00
Andy2
da02e994b7 protext min,max with compile-time flag 2011-11-04 06:16:44 -07:00
Eric House
a2319a8073 experimental additions, mostly compile-time disable-able, so prune
dict by wordlen.
2011-11-03 21:25:51 -07:00
Andy2
5c995af38c stop changing formats between internal and external structs
(iterators); use internal format all the time for speed.  Update linux
test code for new API.
2011-11-02 18:32:00 -07:00
Andy2
eb2b139e48 add some consts and better test for null index data 2011-11-01 18:47:20 -07:00
Andy2
fc4a5c31f4 silence picky compiler 2011-10-31 21:11:52 -07:00
Andy2
0d4404d43b add dict_getStartsWith(), and code to test it. Lots of cleanup and
bugfixing of dictiter code as well.
2011-10-31 21:05:34 -07:00
Andy2
69b1b4da27 break dict iteration stuff out into new files. No code change
otherwise.
2011-10-31 18:30:55 -07:00