X-Git-Url: http://gitweb.fperrin.net/?p=gen-quickdic.git;a=blobdiff_plain;f=GCIDE_to_tab_separated;fp=GCIDE_to_tab_separated;h=99a497392eadf854b980032be5af678e966f7a38;hp=0000000000000000000000000000000000000000;hb=028a6accf7d7d721797482d65dd4dc9d13611840;hpb=5da87a9ec2370bb2f7ce11e107f07625e42f7171

diff --git a/GCIDE_to_tab_separated b/GCIDE_to_tab_separated
new file mode 100755
index 0000000..99a4973
--- /dev/null
+++ b/GCIDE_to_tab_separated
@@ -0,0 +1,173 @@
+#!/usr/bin/python3
+
+import lxml.etree
+import re
+
+entity_map = {
+    "<br/":    "<br/>",
+    "&":       "&amp;",
+    "<lt/":    "&lt;",
+    "<gt/":    "&gt;",
+    "--":      "—", # long (em) dash
+    "<Cced/":  "Ç", # C cedilla
+    "<uum/":   "ü", # u umlaut (diaeresis)
+    "<eacute/":"é", # e acute
+    "<acir/":  "â", # a circumflex
+    "<aum/":   "ä", # a umlaut (diaeresis)
+    "<agrave/":"à", # a grave
+    "<aring/": "Ã¥", # a ring above
+    "<cced/":  "ç", # c cedilla
+    "<ecir/":  "ê", # e circumflex
+    "<eum/":   "ë", # e umlaut (diaeresis)
+    "<egrave/":"è", # e grave
+    "<ium/":   "ï", # i umlaut (diaeresis)
+    "<icir/":  "î", # i circumflex
+    "<icirc/": "î", # i circumflex
+    "<igrave/":"ì", # i grave
+    "<Aum/":   "Ä", # A umlaut
+    "<Eacute/":"É", # E acute
+    "<ae/":    "æ", # ligature ae
+    "<AE/":    "Æ", # ligature AE
+    "<ocir/":  "ô", # o circumflex
+    "<oum/":   "ö", # o umlaut (diaeresis)
+    "<ograve/":"ò", # o grave
+    "<ucir/":  "û", # u circumflex
+    "<ugrave/":"ù", # u grave
+    "<yum/":   "ÿ", # y umlaut
+    "<Oum/":   "Ö", # O umlaut
+    "<Uum/":   "Ü", # U umlaut (diaeresis)
+    "<pound/": "£", # pound sign (British)
+    "<aacute/":"á", # a acute
+    "<iacute/":"í", # i acute
+    "<oacute/":"ó", # o acute
+    "<uacute/":"ú", # u acute
+    "<ntil/":  "ñ", # n tilde
+    "<Ntil/":  "Ñ", # N tilde
+    "<frac23/":"⅔", # two-thirds
+    "<frac13/":"⅓", # one-third
+    "<sec/":   "˝", # seconds (of degree or time). Also, inches or double prime.
+    "<frac12/":"½", # one-half
+    "<frac14/":"¼", # one-quarter
+    "<hand/":  "☞", # pointing hand (printer's "fist")
+    "<bprime/":"˝", # bold accent (used in pronunciations)
+    "<prime/": "´", # light accent (used in pronunciations) also minutes (of
+                    # arc or time)
+    "<rdquo/": "”", # close double quote
+    "<sect/":  "§", # section mark
+    "<ldquo/": "“", # open double quotes
+    "<amac/":  "ā", # a macron
+    "<lsquo/": "‘", # left single quote
+    "<nsm/":   "ṉ", # "n sub-macron"
+    "<sharp/": "♯", # musical sharp
+    "<flat/":  "♭", # musical flat
+    "<imac/":  "Ä«", # i macron
+    "<emac/":  "ē", # e macron
+    "<dsdot/": "ḍ", # Sanskrit/Tamil d dot 
+    "<nsdot/": "ṇ", # Sanskrit/Tamil n dot
+    "<tsdot/": "á¹­", # Sanskrit/Tamil t dot
+    "<ecr/":   "ĕ", # e breve
+    "<icr/":   "Ä­", # i breve
+    "<ocr/":   "ŏ", # o breve
+    "<OE/":    "Œ", # OE ligature
+    "<oe/":    "œ", # oe ligature
+    "<omac/":  "ō", # o macron
+    "<umac/":  "Å«", # u macron
+    "<ocar/":  "ǒ", # o hacek
+    "<aemac/": "Ç£", # ae ligature macron
+    "<oemac/": "ō", # oe ligature macron
+    "<ucr/":   "Å­", # u breve
+    "<acr/":   "ă", # a breve
+    "<cre/":   "˘", # crescent (like a breve, but vertically centered --
+                    # represents the short accent in poetic meter)
+    "<ymac/":  "ȳ", # y macron
+    "<edh/":   "ð", # small eth
+    "<thorn/": "þ", # small thorn
+    "<atil/":  "ã", # a tilde
+    "<ndot/":  "ṅ", # n with dot above
+    "<rsdot/": "ṛ", # r with a dot below
+    "<yogh/":  "ȝ", # small yogh
+    "<mdash/": "—", # em dash
+    "<divide/":"÷", # division sign
+    "<deg/":   "°", # degree sign
+    "<middot/":"•", # bold middle dot
+    "<root/":  "√", # root sign
+    "<adot/":  "ȧ", # a with dot above
+
+    "<?/":     "?", #(?) Place-holder for unknown or illegible character.
+
+    # used only in prononciation key; not able to find what "short vertical
+    # bar on top" looks like with unicode chars.
+    "<asl/":   "a", #  a "semilong" (has a macron above with a short
+                    # vertical bar on top the center of the macron)
+                    # Used in pronunciations.
+    "<esl/":   "e", # e "semilong"
+    "<isl/":   "i", # i "semilong"
+    "<osl/":   "o", # o "semilong"
+    "<usl/":   "u", # u "semilong"
+    "<th/":    "th",# th ligature
+    "<ait/":   "𝑎", # a italic
+    "<eit/":   "𝑒",
+    "<iit/":   "𝑖",
+    "<oit/":   "𝑜",
+    "<uit/":   "𝑢",
+    "<add/":   "a", # a with two dot below
+    "<edd/":   "e",
+    "<idd/":   "i",
+    "<odd/":   "o",
+    "<udd/":   "u",
+    "<oocr/":  "oo",
+    "<oomac/":  "oo",
+    "<etil/":  "ẽ",
+    "<ycr/":   "ў",
+
+    # greek letters
+    "<alpha/": "α",         "<ALPHA/": "Α",
+    "<beta/": "β",          "<BETA/": "Β",
+    "<gamma/": "γ",         "<GAMMA/": "Γ",
+    "<delta/": "δ",         "<DELTA/": "Δ",
+    "<epsilon/": "ε",       "<EPSILON/": "Ε",
+    "<zeta/": "ζ",          "<ZETA/": "Ζ",
+    "<eta/": "η",           "<ETA/": "Η",
+    "<theta/": "θ",         "<THETA/": "Θ",
+    "<iota/": "ι",          "<IOTA/": "Ι",
+    "<kappa/": "κ",         "<KAPPA/": "Κ",
+    "<lambda/": "λ",        "<LAMBDA/": "Λ",
+    "<mu/": "μ",            "<MU/": "Μ",
+    "<nu/": "ν",            "<NU/": "Ν",
+    "<xi/": "ξ",            "<XI/": "Ξ",
+    "<omicron/": "ο",       "<OMICRON/": "Ο",
+    "<pi/": "π",            "<PI/": "Π",
+    "<rho/": "ρ",           "<RHO/": "Ρ",
+    "<sigma/": "σ",         "<SIGMA/": "Σ",
+    "<tau/": "τ",           "<TAU/": "Τ",
+    "<upsilon/": "υ",       "<UPSILON/": "Υ",
+    "<phi/": "φ",           "<PHI/": "Φ",
+    "<chi/": "χ",           "<CHI/": "Χ",
+    "<psi/": "ψ",           "<PSI/": "Ψ",
+    "<omega/": "ω",         "<OMEGA/": "Ω",
+
+    # then there are some characters that are shown as escape sequences
+    r"\'94":   "ö",
+    r"\'d8":   "‖",
+    r"/'bd":   "“",        # one instance where / is used instead of \
+    r" 'bd":   "“",        # two instances where \ is misssing
+    r"`'b8":   "”",        # one instance where ` is used instead of \
+
+    # entities that appear in the etymology of Arabic words, but no explanation
+    # of what they stand for. Not displayed at all by GNU dico.
+    "<hsdot/": "",
+    "<zsdot/": "",
+}
+
+def replace_fake_comments(match):
+    nblines = match.group(0).count("\n")
+    return "\n" * nblines
+
+def convert_file(fname):
+    rawtext = open(fname, "r").read()
+    rawtext = re.sub(r"<--.+?-->", replace_fake_comments, rawtext, flags=re.DOTALL)
+    rawtext = f"<dict>" + rawtext + f"</dict>"
+    for entity, char in entity_map.items():
+        rawtext = rawtext.replace(entity, char)
+    print(rawtext.splitlines()[5724:5730])
+    e = lxml.etree.XML(rawtext)