fcitx 3.2.1补订，曾加三个新工具：readPYMB、readPYBase、mb2org

wpyh · 发表于 2006-7-17 01:58:50

这些工具是为了提取用户的词库（个人词库）而写的。

1、readPYMB，这个程序读取用户词库的内容，然后显示。输出是这样的：

PYFAIndex: 7
HZ: 政
UserPhraseCount: 1
+-Length: 2
| Map: VW
| Phrase: 府
| Index: 32169
| Hit: 8
PYFAIndex: 7
HZ: 正
UserPhraseCount: 2
+-Length: 2
| Map: Jb
| Phrase: 确
| Index: 30425
| Hit: 1
+-Length: 2
| Map: BB
| Phrase: 在
| Index: 36211
| Hit: 3

复制代码

2、readPYBase，这是我要写mb2org的时候，先写的程序，功能是显示pybase.mb文件的内容。部分输出如下：

AX: HZ Index
抓 27661
爪 26844
挝 26046
摣 25265
AY: HZ Index
拽 27660
嘬 26843
尵 26045

复制代码

以上例子AX、AY是Map码。

3、mb2org，是最重要的工具，他的功能是由用户词库.mb文件转化为.org文件（输出到stdout）。例如（部分输出）：

zhi'hou 之后
zhi'jian 之间
zhi'qian 之前
zhi'bu'guo 只不过
zhi'neng 只能
zhi'shi 只是
zhi'you 只有
zhi'yao 只要

复制代码

这些工具的补订如下：（还有，稍微修正了tools/Makefile.am）

--- fcitx-3.2.1/tools/Makefile.am 2006-07-15 16:16:23.000000000 +0800
+++ fcitx-3.2.1/tools/Makefile.am 2006-07-16 23:00:45.000000000 +0800
@@ -1,13 +1,22 @@
+DEFS = -DPKGDATADIR="$(pkgdatadir)"
+
toolsdir = $(pkgdatadir)/tools
INCLUDES = -I$(top_srcdir)/src -I$(top_srcdir)/lib
-bin_PROGRAMS = mb2txt txt2mb createPYMB jd2fcitx pyjj2fcitx win2fcitx
+bin_PROGRAMS = mb2txt txt2mb createPYMB jd2fcitx pyjj2fcitx win2fcitx readPYMB readPYBase mb2org
createPYMB_LDADD = ../src/pyParser.o ../src/pyMapTable.o ../src/PYFA.o ../src/sp.o
pyjj2fcitx_LDADD = ../src/pyParser.o ../src/pyMapTable.o ../src/PYFA.o ../src/sp.o
+readPYMB_LDADD = pyTools.o
+readPYBase_LDADD = pyTools.o
+mb2org_LDADD = ../src/pyParser.o ../src/pyMapTable.o ../src/PYFA.o ../src/sp.o pyTools.o
all:
./txt2mb $(srcdir)/../data/cj.txt ../data/cj.mb
./txt2mb $(srcdir)/../data/erbi.txt ../data/erbi.mb
./txt2mb $(srcdir)/../data/wbx.txt ../data/wbx.mb
- ./createPYMB $(srcdir)/../data/gbkpy.org $(srcdir)/../data/pyPhrase.org && mv *.mb ../data && rm -f pyERROR ; rm -f pyPhrase.ok
+ ./createPYMB $(srcdir)/../data/gbkpy.org $(srcdir)/../data/pyPhrase.org
+ mv *.mb ../data
+ rm -f pyERROR
+ rm -f pyPhrase.ok
+
--- fcitx-3.2.1/tools/mb2org.c 1970-01-01 08:00:00.000000000 +0800
+++ fcitx-3.2.1/tools/mb2org.c 2006-07-17 01:32:45.000000000 +0800
@@ -0,0 +1,113 @@
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+
+#include "pyParser.h"
+#include "pyMapTable.h"
+#include "PYFA.h"
+#include "sp.h"
+#include "pyTools.h"
+
+/* Bad programming practice :( */
+Bool bFullPY;
+Bool bSingleHZMode;
+
+void usage();
+char *HZToPY(struct _HZMap *, char []);
+
+int main(int argc, char **argv)
+{
+ FILE *fi, *fi2;
+ int i, j, k;
+ char *pyusrphrase_mb, *pybase_mb, *HZPY, tMap[3], tPY[10];
+ struct _HZMap *HZMap;
+ struct _PYMB *PYMB;
+
+ if (argc > 3)
+ usage();
+
+ pyusrphrase_mb = getuserfile(PY_USERPHRASE_FILE, (argc > 1) ? argv[1] : "");
+ fi = tryopen(pyusrphrase_mb);
+
+ pybase_mb = strdup((argc > 2) ? argv[2] : (PKGDATADIR "/data/" PY_BASE_FILE));
+ fi2 = tryopen(pybase_mb);
+
+ LoadPYMB(fi, &PYMB);
+ LoadPYBase(fi2, &HZMap);
+
+ for (i = 0; PYMB[i].HZ[0]; ++i)
+ {
+ for (j = 0; j < PYMB[i].UserPhraseCount; ++j)
+ {
+ HZPY = HZToPY(&(HZMap[PYMB[i].PYFAIndex]), PYMB[i].HZ);
+ printf("%s", HZPY);
+
+ for (k = 0; k < PYMB[i].UserPhrase[j].Length / 2; ++k)
+ {
+ memcpy(tMap, PYMB[i].UserPhrase[j].Map + 2 * k, 2);
+ tMap[2] = '\0';
+ tPY[0] = '\0';
+ if (!MapToPY(tMap, tPY))
+ strcpy(tPY, "'*");
+ printf("'%s", tPY);
+ }
+ printf(" %s%s\n", PYMB[i].HZ, PYMB[i].UserPhrase[j].Phrase);
+
+ free(HZPY);
+ }
+ printf("\n");
+ }
+
+ return 0;
+}
+
+/*
+ This function takes a HanZi (HZ) and returns a PinYin (PY) string.
+ If no match is found, "*" is returned.
+*/
+
+char *HZToPY(struct _HZMap *pHZMap1, char HZ[3])
+{
+ int i;
+ char Map[3], tPY[10];
+
+ Map[0] = '\0';
+ for (i = 0; i < pHZMap1->BaseCount; ++i)
+ if (memcmp(HZ, pHZMap1->HZ + 2 * i, 2))
+ {
+ strcpy(Map, pHZMap1->Map);
+ break;
+ }
+
+ if (!Map[0] || !MapToPY(Map, tPY))
+ strcpy(tPY, "*");
+
+ return strdup(tPY);
+}
+
+void usage()
+{
+ puts(
+"mb2org - Convert .mb file to .org file (SEE NOTES BELOW)\n"
+"\n"
+" usage: mb2org [<pyusrphrase.mb>] [<pybase.mb>]\n"
+"\n"
+" <pyusrphrase.mb> this is the .mb file to be decoded, usually this is\n"
+" ~/.fcitx/" PY_USERPHRASE_FILE "\n"
+" if not specified, defaults to\n"
+" ~/.fcitx/" PY_USERPHRASE_FILE "\n"
+" <pybase.mb> this is the pybase.mb file used to determine the\n"
+" of the first character in HZ. Usually, this is\n"
+" " PKGDATADIR "/data/" PY_BASE_FILE "\n"
+" if not specified, defaults to\n"
+" " PKGDATADIR "/data/" PY_BASE_FILE "\n"
+"\n"
+"NOTES:\n"
+"1. If no match is found for a particular HZ, then the pinyin for that HZ\n"
+" will be `*'.\n"
+"2. Always check the produced output for errors.\n"
+ );
+ exit(1);
+ return;
+}
+
--- fcitx-3.2.1/tools/pyTools.c 1970-01-01 08:00:00.000000000 +0800
+++ fcitx-3.2.1/tools/pyTools.c 2006-07-17 01:05:26.000000000 +0800
@@ -0,0 +1,138 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "pyTools.h"
+
+void LoadPYMB(FILE *fi, struct _PYMB **pPYMB)
+{
+ struct _PYMB *PYMB;
+ int i, j, r, n, t, t2;
+
+ /* Is there a way to avoid reading the whole file twice? */
+
+ /* First Pass: Determine the size of the PYMB array to be created */
+
+ n = 0;
+ while (1)
+ {
+ r = fread(&t, sizeof (int), 1, fi);
+ if (!r)
+ break;
+ ++n;
+
+ fseek(fi, 2, SEEK_CUR);
+ fread(&t, sizeof (int), 1, fi);
+
+ for (i = 0; i < t; ++i)
+ {
+ fread(&t2, sizeof (int), 1, fi);
+ fseek(fi, 2 * t2 + 2 * sizeof (int), SEEK_CUR);
+ }
+ }
+
+ /* Second Pass: Actually read the data */
+
+ fseek(fi, 0, SEEK_SET);
+
+ *pPYMB = PYMB = malloc(sizeof (*PYMB) * (n + 1));
+
+ for (i = 0; i < n; ++i)
+ {
+ r = fread(&(PYMB[i].PYFAIndex), sizeof (int), 1, fi);
+
+ fread(PYMB[i].HZ, sizeof (char) * 2, 1, fi);
+ PYMB[i].HZ[2] = '\0';
+
+ fread(&(PYMB[i].UserPhraseCount), sizeof (int), 1, fi);
+ PYMB[i].UserPhrase = malloc(sizeof(*(PYMB[i].UserPhrase)) * PYMB[i].UserPhraseCount);
+
+#define PU(i,j) (PYMB[(i)].UserPhrase[(j)])
+ for (j = 0; j < PYMB[i].UserPhraseCount; ++j)
+ {
+ fread(&(PU(i,j).Length), sizeof (int), 1, fi);
+
+ PU(i,j).Map = malloc(sizeof (char) * PU(i,j).Length + 1);
+ fread(PU(i,j).Map, sizeof (char) * PU(i,j).Length, 1, fi);
+ PU(i,j).Map[PU(i,j).Length] = '\0';
+
+ PU(i,j).Phrase = malloc(sizeof (char) * PU(i,j).Length + 1);
+ fread(PU(i,j).Phrase, sizeof (char) * PU(i,j).Length, 1, fi);
+ PU(i,j).Phrase[PU(i,j).Length] = '\0';
+
+ fread(&(PU(i,j).Index), sizeof (int), 1, fi);
+
+ fread(&(PU(i,j).Hit), sizeof (int), 1, fi);
+ }
+#undef PU
+ }
+ PYMB[n].HZ[0] = '\0';
+
+ return;
+}
+
+int LoadPYBase(FILE *fi, struct _HZMap **pHZMap)
+{
+ int i, j, r, PYFACount;
+ struct _HZMap *HZMap;
+
+ r = fread(&PYFACount, sizeof (int), 1, fi);
+ if (!r)
+ return 0;
+
+ *pHZMap = HZMap = malloc(sizeof (*HZMap) * (PYFACount + 1));
+ for (i = 0; i < PYFACount; ++i)
+ {
+ fread(HZMap[i].Map, 2, 1, fi);
+ HZMap[i].Map[2] = '\0';
+
+ fread(&(HZMap[i].BaseCount), sizeof (int), 1, fi);
+ HZMap[i].HZ = malloc(2 * HZMap[i].BaseCount);
+ HZMap[i].Index = malloc(sizeof (int) * HZMap[i].BaseCount);
+
+ for (j = 0; j < HZMap[i].BaseCount; ++j)
+ {
+ fread(HZMap[i].HZ + j * 2, 2, 1, fi);
+ fread(HZMap[i].Index + j, sizeof (int), 1, fi);
+ }
+ }
+ HZMap[i].Map[0] = '\0';
+
+ return PYFACount;
+}
+
+FILE *tryopen(char *filename)
+{
+ FILE *fi;
+
+ fi = fopen(filename, "r");
+ if (!fi)
+ {
+ perror("fopen");
+ fprintf(stderr, "Can't open file `%s' for reading\n", filename);
+ exit(1);
+ }
+
+ return fi;
+}
+
+char *getuserfile(char *name, char *given)
+{
+ char *filename, *home;
+
+ if (given[0])
+ filename = strdup(given);
+ else
+ {
+ home = getenv("HOME");
+ if (!home)
+ home = strdup("~");
+ filename = malloc(strlen(home) + strlen("/.fcitx/") + strlen(name) + 1);
+ strcpy(filename, home);
+ strcat(filename, "/.fcitx/");
+ strcat(filename, name);
+ }
+
+ return filename;
+}
+
--- fcitx-3.2.1/tools/pyTools.h 1970-01-01 08:00:00.000000000 +0800
+++ fcitx-3.2.1/tools/pyTools.h 2006-07-17 01:05:43.000000000 +0800
@@ -0,0 +1,35 @@
+#ifndef _PY_TOOLS_H
+#define _PY_TOOLS_H
+
+struct _PYMB
+{
+ int PYFAIndex;
+ char HZ[3];
+ int UserPhraseCount;
+ struct
+ {
+ int Length;
+ char *Map;
+ char *Phrase;
+ int Index;
+ int Hit;
+ } *UserPhrase;
+};
+
+struct _HZMap
+{
+ char Map[3];
+ int BaseCount;
+ char *HZ;
+ int *Index;
+};
+
+int LoadPYBase(FILE *, struct _HZMap **);
+void LoadPYMB(FILE *, struct _PYMB **);
+
+char *getuserfile(char *, char *);
+
+FILE *tryopen(char *);
+
+#endif /* _PY_TOOLS_H */
+
--- fcitx-3.2.1/tools/readPYBase.c 1970-01-01 08:00:00.000000000 +0800
+++ fcitx-3.2.1/tools/readPYBase.c 2006-07-17 00:33:25.000000000 +0800
@@ -0,0 +1,66 @@
+#include <stdio.h>
+
+#include "py.h"
+#include "pyTools.h"
+
+void usage();
+
+int main(int argc, char **argv)
+{
+ FILE *fi;
+ int i, PYFACount;
+ char *pybase_mb;
+ struct _HZMap *HZMap;
+
+ if (argc > 2)
+ usage();
+
+ pybase_mb = strdup((argc > 1) ? argv[1] : (PKGDATADIR "/data/" PY_BASE_FILE));
+ fi = tryopen(pybase_mb);
+
+ PYFACount = LoadPYBase(fi, &HZMap);
+ if (PYFACount > 0)
+ {
+#if 0
+ for (i = 0; i < PYFACount; ++i)
+ {
+ printf("%s: ", HZMap[i].Map);
+ fwrite(HZMap[i].HZ, 2, HZMap[i].BaseCount, stdout);
+ printf("\n\n");
+ }
+#else
+ for (i = 0; i < PYFACount; ++i)
+ {
+ int j;
+ printf("%s: HZ Index\n", HZMap[i].Map);
+ for (j = 0; j < HZMap[i].BaseCount / 2; ++j)
+ {
+ printf(" ");
+ fwrite(HZMap[i].HZ + 2 * j, 2, 1, stdout);
+ printf(" %5d\n", *(HZMap[i].Index + 2 * j));
+ }
+ printf("\n");
+ }
+#endif
+ }
+
+ return 0;
+}
+
+void usage()
+{
+ puts(
+"readPYBase - read pybase.mb file and display its contents\n"
+"\n"
+" usage: readPYBase [<pybase.mb>]\n"
+"\n"
+" <pybase.mb> full path to the file, usually\n"
+" " PKGDATADIR "/data/" PY_BASE_FILE "\n"
+" if not specified, defaults to\n"
+" " PKGDATADIR "/data/" PY_BASE_FILE "\n"
+"\n"
+ );
+ exit(1);
+ return;
+}
+
--- fcitx-3.2.1/tools/readPYMB.c 1970-01-01 08:00:00.000000000 +0800
+++ fcitx-3.2.1/tools/readPYMB.c 2006-07-17 01:04:31.000000000 +0800
@@ -0,0 +1,61 @@
+#include <stdio.h>
+#include <stdlib.h>
+
+#include "py.h"
+#include "pyTools.h"
+
+void usage();
+
+int main(int argc, char **argv)
+{
+ FILE *fi;
+ int i, j;
+ char *pyusrphrase_mb;
+ struct _PYMB *PYMB;
+
+ if (argc > 3)
+ usage();
+
+ pyusrphrase_mb = getuserfile(PY_USERPHRASE_FILE, (argc > 1) ? argv[1] : "");
+ fi = tryopen(pyusrphrase_mb);
+ LoadPYMB(fi, &PYMB);
+
+ for (i = 0; PYMB[i].HZ[0]; ++i)
+ {
+ printf("PYFAIndex: %d\n", PYMB[i].PYFAIndex);
+ printf("HZ: %s\n", PYMB[i].HZ);
+ printf("UserPhraseCount: %d\n", PYMB[i].UserPhraseCount);
+
+ for (j = 0; j < PYMB[i].UserPhraseCount; ++j)
+ {
+ printf("+-Length: %d\n", PYMB[i].UserPhrase[j].Length);
+ printf("| Map: %s\n", PYMB[i].UserPhrase[j].Map);
+ printf("| Phrase: %s\n", PYMB[i].UserPhrase[j].Phrase);
+ printf("| Index: %d\n", PYMB[i].UserPhrase[j].Index);
+ printf("| Hit: %d\n", PYMB[i].UserPhrase[j].Hit);
+ }
+ printf("\n");
+ }
+
+ return 0;
+}
+
+void usage()
+{
+ puts(
+"readPYMB - read data from a pinyin .mb file and display its meaning\n"
+"\n"
+" usage: readPYMB <mbfile>\n"
+"\n"
+" <mbfile> MB (MaBiao) file to be read, usually this is\n"
+" ~/.fcitx/" PY_USERPHRASE_FILE "\n"
+" if not specified, defaults to\n"
+" ~/.fcitx/" PY_USERPHRASE_FILE "\n"
+"\n"
+" The MB file can either be a user's MB file (~/.fcitx/pyuserphrase.mb),\n"
+" or the system phrase pinyin MB file (/usr/share/fcitx/data/pyphrase.mb.\n"
+ );
+ exit(1);
+ return;
+}
+

复制代码

我的代码风格和Yuking的不太一样，呵呵~
希望会接受~

wpyh · 发表于 2006-7-17 02:46:48

漏了一个补订（不然的话编译时会出错）：

--- fcitx-3.2.1/Makefile.am 2006-06-06 22:16:37.000000000 +0800
+++ fcitx-3.2.1/Makefile.am 2006-07-17 02:26:37.000000000 +0800
@@ -1,5 +1,5 @@
AUTOMAKE_OPTIONS = foreign
-SUBDIRS = doc tools data debian xpm lib src
+SUBDIRS = doc debian xpm lib src tools data
EXTRA_DIST = autogen.sh fcitx.spec
MAINTAINERCLEANFILES = \

复制代码

Yuking · 发表于 2006-7-17 08:43:53

OK，òÏ 入所有的补丁

		自动登录	找回密码
密码			注册

fcitx 3.2.1补订，曾加三个新工具：readPYMB、readPYBase、mb2org

浏览过的版块