cli: + kanji; fix slow start

Makefile

@@ -3,7 +3,8 @@ SHELL = /bin/bash
 .PHONY: test
 
 test:
-	export PYTHONPATH=$$PWD/src       ;\
-	python3 -m jiten.jmdict --doctest ;\
-	python3 -m jiten.kanji  --doctest ;\
+	export PYTHONPATH=$$PWD/src       ;  \
+	python3 -m jiten.freq   --doctest && \
+	python3 -m jiten.jmdict --doctest && \
+	python3 -m jiten.kanji  --doctest && \
 	python3 -m jiten.misc   --doctest

src/jiten/cli.py

@@ -55,9 +55,6 @@ def jmdict(ctx, lang, word, max, query):                        # {{{1
     click.echo("query: " + click.style(q, fg = "bright_red"))
     click.echo()
   for e, rank in J.search(q, lang, max):
-    if ctx.obj["verbose"]:
-      click.echo("seq# " + click.style(str(e.seq), fg = "blue")
-                 + ", freq# " + click.style(str(rank), fg = "cyan"))
     click.echo(" | ".join(
       click.style(k.elem, fg = "bright_yellow") for k in e.kanji
     ))
@@ -74,16 +71,54 @@ def jmdict(ctx, lang, word, max, query):                        # {{{1
         click.echo(click.style("* ", fg = "magenta") + t)
     if e.usually_kana():
       click.echo("[" + J.USUKANA + "]")
+    if ctx.obj["verbose"]:
+      click.echo("seq# " + click.style(str(e.seq), fg = "blue")
+                 + ", freq# " + click.style(str(rank), fg = "cyan"))
     click.echo()
                                                                 # }}}1
 
 # TODO
 @cli.command(help = "Search KanjiDic.")
-@click.argument("query")
+@click.option("-w", "--word", is_flag = True,
+              help = "Match whole word (same as \\b...\\b).")
+@click.option("-m", "--max", default = None, type = click.INT,
+              help = "Maximum number of results.")
+@click.argument("query", required = False)
 @click.pass_context
-def kanji(ctx, query):                                          # {{{1
-  click.echo(click.style("TODO", fg = "red"))
-  ctx.exit(1)
+def kanji(ctx, word, max, query):                               # {{{1
+  q = query or click.prompt("query")
+  if word: q = "\\b" + q + "\\b"
+  if ctx.obj["verbose"]:
+    click.echo("query: " + click.style(q, fg = "bright_red"))
+    click.echo()
+  for e in K.search(q, max):
+    click.echo(e.char)
+    click.echo(" | ".join(
+      click.style(r, fg = "bright_yellow") for r in e.on
+    ) or "[no on readings]")
+    click.echo(" | ".join(
+      click.style(r, fg = "bright_green") for r in e.kun
+    ) or "[no kun readings]")
+    click.echo(" | ".join(
+      click.style(r, fg = "cyan") for r in e.nanori
+    ) or "[no name readings]")
+    for m in e.meaning:
+      click.echo(click.style("* ", fg = "magenta") + m)
+    if ctx.obj["verbose"]:
+      click.echo(
+        click.style(hex(ord(e.char)), fg = "blue")
+        + ", " + click.style(str(e.strokes), fg = "yellow")
+        + " strokes"
+        + (", grade " + click.style(e.level, fg = "cyan")
+           if e.level else "")
+        + (", freq# " + click.style(str(e.freq), fg = "magenta")
+           if e.freq else "")
+        + (", old jlpt " + click.style(str(e.jlpt), fg = "blue")
+           if e.jlpt else "")
+        + (", skip " + click.style(e.skip, fg = "yellow")
+           if e.skip else "")
+      )
+    click.echo()
                                                                 # }}}1
 
 # TODO

src/jiten/freq.py

@@ -80,10 +80,22 @@ def rank_freq(data):
 
 def rank(w): return freq_rank.get(w, NOFREQ)
 
-news_freq = process_freq(parse_freq(NEWSFREQ_FILE, True))
-book_freq = process_freq(parse_freq(BOOKFREQ_FILE, False))
-
-freq      = merge_freq(news_freq, book_freq)
-freq_rank = rank_freq(freq)
+def setup():
+  global news_freq, book_freq, freq, freq_rank
+  if setup.done: return
+  setup.done = True
+
+  news_freq = process_freq(parse_freq(NEWSFREQ_FILE, True))
+  book_freq = process_freq(parse_freq(BOOKFREQ_FILE, False))
+
+  freq      = merge_freq(news_freq, book_freq)
+  freq_rank = rank_freq(freq)
+setup.done = False
+
+if __name__ == "__main__":
+  if "--doctest" in sys.argv:
+    setup()
+    import doctest
+    if doctest.testmod(verbose = True)[0]: sys.exit(1)
 
 # vim: set tw=70 sw=2 sts=2 et fdm=marker :

src/jiten/jmdict.py

@@ -292,13 +292,14 @@ def jmdict2sqldb(data, file = SQLITE_FILE):                     # {{{1
 """                                                             # }}}1
 
 def setup():
+  F.setup()
   jmdict = parse_jmdict()
   jmdict2sqldb(jmdict)
 
 def search(q, langs = [DLANG], max_results = None,              # {{{1
            file = SQLITE_FILE):
   entries = set()
-  rx      = re.compile(q, re.I)
+  rx      = re.compile(q, re.I | re.M)
   mat     = lambda x: rx.search(x) is not None
   with sqlite_do(file) as c:
     c.connection.create_function("matches", 1, mat)

src/jiten/kanji.py

@@ -175,9 +175,32 @@ def setup():
   kanjidic = parse_kanjidic()
   kanjidic2sqldb(kanjidic)
 
-# TODO
 def search(q, max_results = None, file = SQLITE_FILE):          # {{{1
-  ...
+  ent   = lambda r: Entry(*(list(r[1:8]) + [ tuple(x.splitlines())
+                                             for x in r[8:] ]))
+  ideo  = tuple(M.uniq(filter(M.isideo, q)))
+  with sqlite_do(file) as c:
+    if ideo:
+      for char in ideo:
+        for i, r in enumerate(c.execute("SELECT * FROM entry WHERE char = ?",
+                                        (char,))):
+          if max_results and i >= max_results: break
+          yield ent(r)
+    else:
+      rx    = re.compile(q, re.I | re.M)
+      mat1  = lambda x: rx.search(x) is not None
+      mat2  = lambda x: rx.search(x.replace(".", "")
+                                   .replace("-", "")) is not None
+      c.connection.create_function("matches1", 1, mat1)
+      c.connection.create_function("matches2", 1, mat2)
+      for i, r in enumerate(c.execute("""
+          SELECT * FROM entry WHERE
+            matches1(on_) OR matches1(kun) OR matches1(nanori) OR
+            matches2(on_) OR matches2(kun) OR matches2(nanori) OR
+            matches1(meaning)
+          """)):
+        if max_results and i >= max_results: break
+        yield ent(r)
                                                                 # }}}1
 
 if __name__ == "__main__":

src/jiten/misc.py

@@ -27,12 +27,14 @@
 ispunc      = lambda c:  0x3000 <= ord(c) <=  0x303f
 ishiragana  = lambda c:  0x3040 <= ord(c) <=  0x309f
 iskatakana  = lambda c:  0x30a0 <= ord(c) <=  0x30ff
+
 iskanji     = lambda c:  0x4e00 <= ord(c) <=  0x9faf
 iscompat    = lambda c:  0xf900 <= ord(c) <=  0xfaff
 isuniext    = lambda c:  0x3400 <= ord(c) <=  0x4dbf or \
                         0x20000 <= ord(c) <= 0x2ebef
 
 iskana      = lambda c: ishiragana(c) or iskatakana(c)
+isideo      = lambda c: iskanji(c) or iscompat(c) or isuniext(c)
 isjap       = lambda c: iskanji(c) or iskana(c)  # probably
 isokjap     = lambda c: isjap(c) or c in OKPUNC  # probably