Mercurial > dottes
view abcfield.py @ 588:afc031477784
Replace sed substitution with Python templating for HTML and LaTeX output.
author | Jim Hague <jim.hague@acm.org> |
---|---|
date | Wed, 02 Nov 2016 00:21:18 +0000 (2016-11-02) |
parents | daa3b76bd11f |
children | 2face6618bf3 |
line wrap: on
line source
#!/usr/bin/env python3 # # Extact a text field (title, by default) from a .abc file, and print it out # with any ABC accented characters converted to HTML (default) or Latex. # # Optionally rearrange a field into display format: # * In Title fields, change 'sort' form such as 'Exploding Potato, The' # to display format 'The Exploding Potato'. # * In Key fields, translate the ABC key representation to full text, # e.g. G#dor becomes G# Dorian. # # Recognise continuation header fields and print those too. The ABC standard # defines continuation fields as starting ':+'. Regrettably none of the tools # I am using the Booke recognise that syntax, so I am adopting a Booke # convention of '<header>:+' *also* being a continuation. Note that a # continuation is a distinct line in the field value; the value has a line # break between it and the previous line. # import optparse import sys accentedletters = { # Acute accents "'A" : ("Á", "\\'{A}"), "'E" : ("É", "\\'{E}"), "'I" : ("Í", "\\'{I}"), "'O" : ("Ó", "\\'{O}"), "'U" : ("Ú", "\\'{U}"), "'Y" : ("Ý", "\\'{Y}"), "'a" : ("á", "\\'{a}"), "'e" : ("é", "\\'{e}"), "'i" : ("í", "\\'{i}"), "'o" : ("ó", "\\'{o}"), "'u" : ("ú", "\\'{u}"), "'y" : ("ý", "\\'{y}"), # Grave accents "`A" : ("À", "\\`{A}"), "`E" : ("È", "\\`{E}"), "`I" : ("Ì", "\\`{I}"), "`O" : ("Ò", "\\`{O}"), "`U" : ("Ù", "\\`{U}"), "`a" : ("à", "\\`{a}"), "`e" : ("è", "\\`{e}"), "`i" : ("ì", "\\`{i}"), "`o" : ("ò", "\\`{o}"), "`u" : ("ù", "\\`{u}"), # Umlauts "\"A" : ("Ä", "\\\"{A}"), "\"E" : ("Ë", "\\\"{E}"), "\"I" : ("Ï", "\\\"{I}"), "\"O" : ("Ö", "\\\"{O}"), "\"U" : ("Ü", "\\\"{U}"), "\"Y" : ("Ÿ", "\\\"{Y}"), "\"a" : ("ä", "\\\"{a}"), "\"e" : ("ë", "\\\"{e}"), "\"i" : ("ï", "\\\"{\i}"), "\"o" : ("ö", "\\\"{o}"), "\"u" : ("ü", "\\\"{u}"), "\"y" : ("ÿ", "\\\"{y}"), # Circumflexes "^A" : ("Â", "\\^{A}"), "^E" : ("Ê", "\\^{E}"), "^I" : ("Î", "\\^{I}"), "^O" : ("Ô", "\\^{O}"), "^U" : ("Û", "\\^{U}"), "^a" : ("â", "\\^{a}"), "^e" : ("ê", "\\^{e}"), "^i" : ("î", "\\^{\i}"), "^o" : ("ô", "\\^{o}"), "^u" : ("û", "\\^{u}"), # Tilde "~A" : ("Ã", "\\~{A}"), "~N" : ("Ñ", "\\~{N}"), "~O" : ("Õ", "\\~{O}"), "~a" : ("ã", "\\~{a}"), "~n" : ("ñ", "\\~{n}"), "~o" : ("õ", "\\~{o}"), # Cedilla ",C" : ("Ç", "\\c{C}"), ",c" : ("ç", "\\c{c}"), # Slash "/O" : ("Ø", "\\O"), "/o" : ("ø", "\\o"), # Ring "AA" : ("Å", "\\r{A}"), "aa" : ("å", "\\r{a}"), # Ligatures "AE" : ("Æ", "\\AE"), "ae" : ("æ", "\\ae"), "ss" : ("ß", "\\ss"), } abckeys = { "m": "Minor", "min": "Minor", "mix": "Mixolydian", "dor": "Dorian", "phr": "Phrygian", "lyd": "Lydian", "loc": "Locrian", } # Convert ABC accented chars to HTML entities or LaTex. def convertAccents(t, latex=False): res = "" while True: p = t.partition('\\') res += p[0] if p[1] == "": break abc = p[2][0:2] t = p[2][2:] if abc in accentedletters: if latex: res += accentedletters[abc][1] else: res += accentedletters[abc][0] else: res += "\\" + abc return res # Convert Title fields from sort to display, so Bat, The->The Bat. def convertTitleToDisplay(t): p = t.rpartition(',') if p[1] == "": return t else: return p[2].strip() + " " + p[0].strip() # Convert Key field from ABC to display, so G#dor->G# Dorian. def convertKeyToDisplay(t): letter = t[0].upper() accidental = "" mode = "" try: accidental = t[1] if accidental == '#' or accidental == 'b': mode = t[2:] else: accidental = "" mode = t[1:] except IndexError: pass mode = mode.strip().lower() return letter + accidental + ' ' + abckeys.get(mode, "Major") # Return the raw text for a given field. Optionally the nth field is taken, # or the field data must start with a designated string to be recognised. def getFieldText(inf, field, n = 1, starts = None): res = "" for line in inf: line = line.strip() if len(line) > 2 and line[1] == ':': if line[0] == "+" or (line[0] == field and line[2] == "+"): if not res: continue if line[0] == "+": line = line[2:] else: line = line[3:] res = res + '\n' + line.strip() else: if res: break if line[0] == field: line = line[2:].strip() if starts: if line.find(starts) != 0: continue line = line[len(starts):].strip() if n > 1: n = n - 1 continue res = line return res # Return display text for a given field. def getFieldDisplayText(inf, field, n = 1, starts = None, latex = False): res = getFieldText(inf, field, n, starts) if res: if field.upper() == "T": res = convertTitleToDisplay(res) elif field.upper() == "K": res = convertKeyToDisplay(res) res = convertAccents(res, latex) return res if __name__ == "__main__": def process(inf, options): if options.display: line = getFieldDisplayText(inf, options.field, options.index, options.starts, options.latex) else: line = getFieldText(inf, options.field, options.index, options.starts) if line: print(line) return True else: return False # execute only if run as a script parser = optparse.OptionParser(usage="usage: %prog [options] [filename]\n\n" " Extract field data from ABC file.") parser.add_option("-f", "--field", dest="field", default="T", help="extract the field FIELD", metavar="FIELD") parser.add_option("-l", "--latex", dest="latex", action="store_true", default=False, help="convert special characters for LaTeX") parser.add_option("-d", "--display", dest="display", action="store_true", default=False, help="convert to display text") parser.add_option("-n", "--index", dest="index", action="store", type="int", default=1, help="report INDEXth value [default: %default]", metavar="INDEX") parser.add_option("-s", "--starts", dest="starts", action="store", type="string", default=None, help="report only if line starts CONTENT and remove CONTENT", metavar="CONTENT") (options, args) = parser.parse_args() res = False if len(args) > 0: for arg in args: try: inf = open(arg, "r") res = res or process(inf, options) finally: inf.close() else: res = process(sys.stdin, options) sys.exit(int(not res))