aboutsummaryrefslogtreecommitdiffhomepage
path: root/share/tools/create_manpage_completions.py
blob: 4b0c2657b1e8a7327d2fdd27d60a41cf1a6c391e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
#!/usr/bin/env python
# -*- coding: utf-8 -*-

# Run me like this: ./create_manpage_completions.py /usr/share/man/man1/* > man_completions.fish

"""
<OWNER> = Siteshwar Vashisht 
<YEAR> = 2012 

Copyright (c) 2012, Siteshwar Vashisht 
All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
"""

import string, sys, re, os.path, gzip, traceback, getopt, errno
from deroff import Deroffer

# Whether we're Python 3
IS_PY3 = sys.version_info[0] >= 3

# This gets set to the name of the command that we are currently executing
CMDNAME = ""

# Information used to track which of our parsers were successful
PARSER_INFO = {}

# builtcommand writes into this global variable, yuck
built_command_output = []

# Diagnostic output
diagnostic_output = []
diagnostic_indent = 0

# Three diagnostic verbosity levels
VERY_VERBOSE, BRIEF_VERBOSE, NOT_VERBOSE = 2, 1, 0

# Pick some reasonable default values for settings
global VERBOSITY, WRITE_TO_STDOUT, DEROFF_ONLY
VERBOSITY, WRITE_TO_STDOUT, DEROFF_ONLY = NOT_VERBOSE, False, False

def add_diagnostic(dgn, msg_verbosity = VERY_VERBOSE):
    # Add a diagnostic message, if msg_verbosity <= VERBOSITY
    if msg_verbosity <= VERBOSITY:
        diagnostic_output.append('   '*diagnostic_indent + dgn)
    
def flush_diagnostics(where):
    if diagnostic_output:
        output_str = '\n'.join(diagnostic_output) + '\n'
        where.write(output_str)
        diagnostic_output[:] = []

# Make sure we don't output the same completion multiple times, which can happen
# For example, xsubpp.1.gz and xsubpp5.10.1.gz
# This maps commands to lists of completions
already_output_completions = {}

def compileAndSearch(regex, input):
    options_section_regex = re.compile(regex , re.DOTALL)
    options_section_matched = re.search( options_section_regex, input)
    return options_section_matched

def unquoteDoubleQuotes(data):
    if (len(data) < 2):
        return data
    if data[0] == '"' and data[len(data)-1] == '"':
        data = data[1:len(data)-1]
    return data

def unquoteSingleQuotes(data):
    if (len(data) < 2):
        return data
    if data[0] == '`' and data[len(data)-1] == '\'':
        data = data[1:len(data)-1]
    return data
    

# Make a string of characters that are deemed safe in fish without needing to be escaped
# Note that space is not included
g_fish_safe_chars = frozenset(string.ascii_letters + string.digits + '_+-|/:=@~')

def fish_escape_single_quote(str):
    # Escape a string if necessary so that it can be put in single quotes
    # If it has no non-safe chars, there's nothing to do
    if g_fish_safe_chars.issuperset(str):
        return str
    
    str = str.replace('\\', '\\\\') # Replace one backslash with two
    str = str.replace("'", "\\'") # Replace one single quote with a backslash-single-quote
    return "'" + str + "'"

def output_complete_command(cmdname, args, description, output_list):
    comps = ['complete -c', cmdname]
    comps.extend(args)
    if description:
        comps.append('--description')
        comps.append(description)
    output_list.append(' '.join(comps))

def builtcommand(options, description):
#    print "Options are: ", options
    man_optionlist = re.split(" |,|\"|=|[|]", options)
    fish_options = []
    for option in man_optionlist:
        option = option.strip()
        
        # Skip some problematic cases
        if option in ['-', '--']: continue
        
        if option.startswith('--'):
            # New style long option (--recursive)
            fish_options.append('-l ' + fish_escape_single_quote(option[2:]))
        elif option.startswith('-') and len(option) == 2:
            # New style short option (-r)
            fish_options.append('-s ' + fish_escape_single_quote(option[1:]))
        elif option.startswith('-') and len(option) > 2:
            # Old style long option (-recursive)
            fish_options.append('-o ' + fish_escape_single_quote(option[1:]))
    
    # Determine which options are new (not already in existing_options)
    # Then add those to the existing options
    existing_options = already_output_completions.setdefault(CMDNAME, set())
    fish_options = [opt for opt in fish_options if opt not in existing_options]
    existing_options.update(fish_options)
    
    # Maybe it's all for naught
    if not fish_options: return
    
    first_period = description.find(".") 
    if first_period >= 45 or (first_period == -1 and len(description) > 45):
        description = description[:45] + '... [See Man Page]'
    elif first_period >= 0:
        description = description[:first_period]
    
    # Escape some more things
    description = fish_escape_single_quote(description)
    escaped_cmd = fish_escape_single_quote(CMDNAME)
    
    output_complete_command(escaped_cmd, fish_options, description, built_command_output)

    

def removeGroffFormatting(data):
#    data = data.replace("\fI","")
#    data = data.replace("\fP","")
    data = data.replace("\\fI","")
    data = data.replace("\\fP","")
    data = data.replace("\\f1","")
    data = data.replace("\\fB","")
    data = data.replace("\\fR","")
    data = data.replace("\\e","")
    data = re.sub(".PD( \d+)","",data)
    data = data.replace(".BI","")
    data = data.replace(".BR","")
    data = data.replace("0.5i","")
    data = data.replace(".rb","")
    data = data.replace("\\^","")
    data = data.replace("{ ","")
    data = data.replace(" }","")
    data = data.replace("\ ","")
    data = data.replace("\-","-")
    data = data.replace("\&","")
    data = data.replace(".B","")
    data = data.replace("\-","-")
    data = data.replace(".I","")
    data = data.replace("\f","")
    return data

class ManParser:
    def isMyType(self, manpage):
        return False

    def parseManPage(self, manpage):
        return False

    def name(self):
        return "no-name"

class Type1ManParser(ManParser):
    def isMyType(self, manpage):
        #   print manpage
        options_section_matched = compileAndSearch("\.SH \"OPTIONS\"(.*?)", manpage)

        if options_section_matched == None:
            return False
        else:
            return True

    def parseManPage(self, manpage):
        options_section_regex = re.compile( "\.SH \"OPTIONS\"(.*?)(\.SH|\Z)", re.DOTALL)
        options_section_matched = re.search( options_section_regex, manpage)
        
        options_section = options_section_matched.group(0)
        #   print options_section
        options_parts_regex = re.compile("\.PP(.*?)\.RE", re.DOTALL)
        options_matched = re.search(options_parts_regex, options_section)
        #   print options_matched
        add_diagnostic('Command is ' + CMDNAME)

        if options_matched == None:
            add_diagnostic('Unable to find options')
            if( self.fallback(options_section) ):
                return True
            elif (self.fallback2(options_section) ):
                return True
            return False
        
        while (options_matched != None):
            #       print len(options_matched.groups())
            #       print options_matched.group()
            data = options_matched.group(1)
            last_dotpp_index = data.rfind(".PP")
            if (last_dotpp_index != -1):
                data = data[last_dotpp_index+3:]

            data = removeGroffFormatting(data)
            data = data.split(".RS 4")
            #       print data
            if (len (data) > 1): #and len(data[1]) <= 300):
                optionName = data[0].strip()
                
                if ( optionName.find("-") == -1):
                    add_diagnostic(optionName + " doesn't contain - ")
#                    return False
                else:
                    optionName = unquoteDoubleQuotes(optionName)
                    optionName = unquoteSingleQuotes(optionName)
                    optionDescription = data[1].strip().replace("\n"," ")
#                    print >> sys.stderr, "Option: ", optionName," Description: ", optionDescription , '\n'
                    builtcommand(optionName, optionDescription)
                    
            else:
                add_diagnostic('Unable to split option from description')
                return False

            options_section = options_section[options_matched.end()-3:]
            options_matched = re.search(options_parts_regex, options_section)

    def fallback(self, options_section):
        add_diagnostic('Falling Back')
        options_parts_regex = re.compile("\.TP( \d+)?(.*?)\.TP", re.DOTALL)
        options_matched = re.search(options_parts_regex, options_section)
        if options_matched == None:
            add_diagnostic('Still not found')
            return False
        while options_matched != None:
            data = options_matched.group(2)
            data = removeGroffFormatting(data)
            data = data.strip()
            data = data.split("\n",1)
            if (len(data)>1 and len(data[1].strip())>0): # and len(data[1])<400):
                optionName = data[0].strip()
                if ( optionName.find("-") == -1):
                    add_diagnostic(optionName + "doesn't contains -")
                else:
                    optionName = unquoteDoubleQuotes(optionName)
                    optionName = unquoteSingleQuotes(optionName)
                    optionDescription = data[1].strip().replace("\n"," ")
#                    print "Option: ", optionName," Description: ", optionDescription , '\n'
                    builtcommand(optionName, optionDescription)
            else:
                add_diagnostic('Unable to split option from description')
                return False

            options_section = options_section[options_matched.end()-3:]
            options_matched = re.search(options_parts_regex, options_section)
        return True

    def fallback2(self, options_section):
        add_diagnostic('Falling Back2')
        ix_remover_regex = re.compile("\.IX.*")
        trailing_num_regex = re.compile('\\d+$')
        options_parts_regex = re.compile("\.IP (.*?)\.IP", re.DOTALL)

        options_section = re.sub(ix_remover_regex, "", options_section)
        options_matched = re.search(options_parts_regex, options_section)
        if options_matched == None:
            add_diagnostic('Still not found2')
            return False
        while options_matched != None:
            data = options_matched.group(1)
            
#            print "Data is : ", data
            data = removeGroffFormatting(data)
            data = data.strip()
            data = data.split("\n",1)
            if (len(data)>1 and len(data[1].strip())>0): # and len(data[1])<400):
#                print "Data[0] is: ", data[0]

#                data = re.sub(trailing_num_regex, "", data)
                optionName = re.sub(trailing_num_regex, "", data[0].strip())
                
                if ('-' not in optionName):
                    add_diagnostic(optionName + " doesn't contain -")
                else:
                    optionName = optionName.strip()
                    optionName = unquoteDoubleQuotes(optionName)
                    optionName = unquoteSingleQuotes(optionName)
                    optionDescription = data[1].strip().replace("\n"," ")
#                    print "Option: ", optionName," Description: ", optionDescription , '\n'
                    builtcommand(optionName, optionDescription)
            else:
#                print data
                add_diagnostic('Unable to split option from description')
                return False

            options_section = options_section[options_matched.end()-3:]
            options_matched = re.search(options_parts_regex, options_section)
        return True
        
    def name(self):
        return "Type1"


class Type2ManParser(ManParser):
    def isMyType(self, manpage):
        options_section_matched = compileAndSearch("\.SH OPTIONS(.*?)", manpage)

        if options_section_matched == None:
            return False
        else:
            return True

    def parseManPage(self, manpage):
        options_section_regex = re.compile( "\.SH OPTIONS(.*?)(\.SH|\Z)", re.DOTALL)
        options_section_matched = re.search( options_section_regex, manpage)
        
#        if (options_section_matched == None):
#            print "Falling Back"
#            options_section_regex = re.compile( "\.SH OPTIONS(.*?)$", re.DOTALL)
#            options_section_matched = re.search( options_section_regex, manpage)
#        print manpage
        options_section = options_section_matched.group(1)
#        print options_section
        #   print options_section
        #   sys.exit(1)

#        options_parts_regex = re.compile("\.TP(.*?)\.TP", re.DOTALL)
        options_parts_regex = re.compile("\.[I|T]P( \d+(\.\d)?i?)?(.*?)\.[I|T]P", re.DOTALL)
#        options_parts_regex = re.compile("\.TP(.*?)[(\.TP)|(\.SH)]", re.DOTALL)
        options_matched = re.search(options_parts_regex, options_section)
        add_diagnostic('Command is ' + CMDNAME)

        if options_matched == None:
            add_diagnostic(self.name() + ': Unable to find options')
            return False

        while (options_matched != None):
            #       print len(options_matched.groups())
            data = options_matched.group(3)

            data = removeGroffFormatting(data)

            data = data.strip()

            data = data.split("\n",1)
#            print >> sys.stderr, data
            if (len(data)>1 and len(data[1].strip())>0): # and len(data[1])<400):
                optionName = data[0].strip()
                if '-' not in optionName:
                    add_diagnostic(optionName + " doesn't contain -")
                else:
                    optionName = unquoteDoubleQuotes(optionName)
                    optionName = unquoteSingleQuotes(optionName)
                    optionDescription = data[1].strip().replace("\n"," ")
#                    print "Option: ", optionName," Description: ", optionDescription , '\n'
                    builtcommand(optionName, optionDescription)
            else:
                # print >> sys.stderr, data
                add_diagnostic('Unable to split option from description')
                
#                return False

            options_section = options_section[options_matched.end()-3:]
            options_matched = re.search(options_parts_regex, options_section)

    

    def name(self):
        return "Type2"


class Type3ManParser(ManParser):
    def isMyType(self, manpage):
        options_section_matched = compileAndSearch("\.SH DESCRIPTION(.*?)", manpage)

        if options_section_matched == None:
            return False
        else:
            return True

    def parseManPage(self, manpage):
        options_section_regex = re.compile( "\.SH DESCRIPTION(.*?)(\.SH|\Z)", re.DOTALL)
        options_section_matched = re.search( options_section_regex, manpage)
        
        options_section = options_section_matched.group(1)
        #   print options_section
        #   sys.exit(1)
        options_parts_regex = re.compile("\.TP(.*?)\.TP", re.DOTALL)
        options_matched = re.search(options_parts_regex, options_section)
        add_diagnostic('Command is ' + CMDNAME)

        if options_matched == None:
            add_diagnostic('Unable to find options section')
            return False

        while (options_matched != None):
#            print len(options_matched.groups())
            data = options_matched.group(1)

            data = removeGroffFormatting(data)
            data = data.strip()
            data = data.split("\n",1)

            if (len(data)>1): # and len(data[1])<400):
                optionName = data[0].strip()
                if ( optionName.find("-") == -1):
                    add_diagnostic(optionName + "doesn't contain -")
                else:
                    optionName = unquoteDoubleQuotes(optionName)
                    optionName = unquoteSingleQuotes(optionName)
                    optionDescription = data[1].strip().replace("\n"," ")
#                    print >> sys.stderr, "Option: ", optionName," Description: ", optionDescription , '\n'
                    builtcommand(optionName, optionDescription)

            else:
                add_diagnostic('Unable to split option from description')
                return False

            options_section = options_section[options_matched.end()-3:]
            options_matched = re.search(options_parts_regex, options_section)



    def name(self):
        return "Type3"


class Type4ManParser(ManParser):
    def isMyType(self, manpage):
        options_section_matched = compileAndSearch("\.SH FUNCTION LETTERS(.*?)", manpage)

        if options_section_matched == None:
            return False
        else:
            return True

    def parseManPage(self, manpage):
        options_section_regex = re.compile( "\.SH FUNCTION LETTERS(.*?)(\.SH|\Z)", re.DOTALL)
        options_section_matched = re.search( options_section_regex, manpage)
        
        options_section = options_section_matched.group(1)
        #   print options_section
        #   sys.exit(1)
        options_parts_regex = re.compile("\.TP(.*?)\.TP", re.DOTALL)
        options_matched = re.search(options_parts_regex, options_section)
        add_diagnostic('Command is ' + CMDNAME)

        if options_matched == None:
            print >> sys.stderr, "Unable to find options section"
            return False

        while (options_matched != None):
            #       print len(options_matched.groups())
            data = options_matched.group(1)

            data = removeGroffFormatting(data)
            data = data.strip()
            data = data.split("\n",1)

            if (len(data)>1): # and len(data[1])<400):
                optionName = data[0].strip()
                if ( optionName.find("-") == -1):
                    add_diagnostic(optionName + " doesn't contain - ")
                else:
                    optionName = unquoteDoubleQuotes(optionName)
                    optionName = unquoteSingleQuotes(optionName)
                    optionDescription = data[1].strip().replace("\n"," ")
#                    print "Option: ", optionName," Description: ", optionDescription , '\n'
                    builtcommand(optionName, optionDescription)

            else:
                add_diagnostic('Unable to split option from description')
                return False

            options_section = options_section[options_matched.end()-3:]
            options_matched = re.search(options_parts_regex, options_section)

        return True

    def name(self):
        return "Type4"

class TypeDarwinManParser(ManParser):
    def isMyType(self, manpage):
        options_section_matched = compileAndSearch("\.S[hH] DESCRIPTION", manpage)
        return options_section_matched != None
    
    def trim_groff(self, line):
        # Remove initial period
        if line.startswith('.'):
            line = line[1:]
        # Skip leading groff crud
        while re.match('[A-Z][a-z]\s', line):
            line = line[3:]
        return line
    
    def is_option(self, line):
        return line.startswith('.It Fl')
    
    
    def parseManPage(self, manpage):
        got_something = False
        lines =  manpage.splitlines()
        # Discard lines until we get to ".sh Description"
        while lines and not (lines[0].startswith('.Sh DESCRIPTION') or lines[0].startswith('.SH DESCRIPTION')):
            lines.pop(0)
        
        while lines:
            # Pop until we get to the next option
            while lines and not self.is_option(lines[0]):
                lines.pop(0)
                
            if not lines:
                continue
            
            # Extract the name
            name = self.trim_groff(lines.pop(0)).strip()
            if not name: continue
            name = name.split(None, 2)[0]
            
            # Extract the description
            desc = ''
            while lines and not self.is_option(lines[0]):
                # print "*", lines[0]
                desc = desc + lines.pop(0)
            
            # print "name: ", name
            
            if name == '-':
                # Skip double -- arguments
                continue
            elif len(name) > 1:
                # Output the command
                builtcommand('--' + name, desc)
                got_something = True
            elif len(name) == 1:
                builtcommand('-' + name, desc)
                got_something = True
                
        return got_something
                
    def name(self):
        return "Darwin man parser"


class TypeDeroffManParser(ManParser):
    def isMyType(self, manpage):
        return True # We're optimists
    
    def is_option(self, line):
        return line.startswith('-')
        
    def could_be_description(self, line):
        return len(line) > 0 and not line.startswith('-')
    
    def parseManPage(self, manpage):
        d = Deroffer()
        d.deroff(manpage)
        output = d.get_output()
        lines = output.split('\n')
        
        got_something = False
        
        # Discard lines until we get to DESCRIPTION or OPTIONS
        while lines and not (lines[0].startswith('DESCRIPTION') or lines[0].startswith('OPTIONS') or lines[0].startswith('COMMAND OPTIONS')):
            lines.pop(0)
        
        # Look for BUGS and stop there
        for idx in range(len(lines)):
            line = lines[idx]
            if line.startswith('BUGS'):
                # Drop remaining elements
                lines[idx:] = []
                break
        
        while lines:
            # Pop until we get to the next option
            while lines and not self.is_option(lines[0]):
                line = lines.pop(0)
            
            if not lines:
                continue
            
            options = lines.pop(0)
            
            # Pop until we get to either an empty line or a line starting with -
            description = ''
            while lines and self.could_be_description(lines[0]):
                if description: description += ' '
                description += lines.pop(0)
            
            builtcommand(options, description)
            got_something = True
            
        return got_something
            
                
    def name(self):
        return "Deroffing man parser"
        
# Return whether the file at the given path is overwritable
# Raises IOError if it cannot be opened
def file_is_overwritable(path):
    result = False
    file = open(path, 'r')
    for line in file:
        # Skip leading empty lines
        line = line.strip()
        if not line:
            continue
        
        # We look in the initial run of lines that start with #
        if not line.startswith('#'):
            break
            
        # See if this contains the magic word
        if 'Autogenerated' in line:
            result = True
            break
            
    file.close()
    return result


# Return whether the file at the given path either does not exist, or exists but appears to be a file we output (and hence can overwrite)
def file_missing_or_overwritable(path):
    try:
        return file_is_overwritable(path)
    except IOError as err:
        if err.errno == 2:
            # File does not exist, full steam ahead
            return True
        else:
            # Something else happened
            return False
    
# Delete the file if it is autogenerated
def cleanup_autogenerated_file(path):
    try:
        if file_is_overwritable(path):
            os.remove(path)
    except (OSError, IOError):
        pass

def parse_manpage_at_path(manpage_path, yield_to_dirs, output_directory):
    filename = os.path.basename(manpage_path)

    # Clear diagnostics
    global diagnostic_indent
    diagnostic_output[:] = []
    diagnostic_indent = 0

    # Set up some diagnostics
    add_diagnostic('Considering ' + manpage_path)
    diagnostic_indent += 1
    
    if manpage_path.endswith('.gz'):
        fd = gzip.open(manpage_path, 'r')
        manpage = fd.read()
        if IS_PY3: manpage = manpage.decode('latin-1')
    else:
        if IS_PY3:
            fd = open(manpage_path, 'r', encoding='latin-1')
        else:
            fd = open(manpage_path, 'r')
        manpage = fd.read()
    fd.close()

    manpage = str(manpage)
    
    # Get the "base" command, e.g. gcc.1.gz -> gcc
    cmd_base = CMDNAME.split('.', 1)[0]
    ignoredcommands = ["cc", "g++", "gcc", "c++", "cpp", "emacs", "gprof", "wget", "ld", "awk"]
    if cmd_base in ignoredcommands:
        return
        
    # Ignore perl's gazillion man pages
    ignored_prefixes = ['perl', 'zsh']
    for prefix in ignored_prefixes:
        if cmd_base.startswith(prefix):
            return
            
    # Ignore the millions of links to BUILTIN(1)
    if manpage.find('BUILTIN 1') != -1:
        return
    
    # Clear the output list
    built_command_output[:] = []
    
    if DEROFF_ONLY:
        parsers = [TypeDeroffManParser()]
    else:
        parsers = [Type1ManParser(), Type2ManParser(), Type4ManParser(), Type3ManParser(), TypeDarwinManParser(), TypeDeroffManParser()]
    parsersToTry = [p for p in parsers if p.isMyType(manpage)]
    
    success = False
    if not parsersToTry:
        add_diagnostic(manpage_path + ": Not supported")
    else:
        for parser in parsersToTry:
            parser_name = parser.name()
            add_diagnostic('Trying parser ' + parser_name)
            diagnostic_indent += 1
            success = parser.parseManPage(manpage)
            diagnostic_indent -= 1
            if success:
                PARSER_INFO.setdefault(parser_name, []).append(CMDNAME)
                break
        
        if success:
            if WRITE_TO_STDOUT:
                output_file = sys.stdout
            else:
                fullpath = os.path.join(output_directory, CMDNAME + '.fish')
                try:
                    if file_missing_or_overwritable(fullpath):
                        output_file = open(fullpath, 'w')
                    else:
                        add_diagnostic("Not overwriting the file at '%s'" % fullpath)
                        
                except IOError as err:
                    add_diagnostic("Unable to open file '%s': error(%d): %s" % (fullpath, err.errno, err.strerror))
                    return False
            
            built_command_output.insert(0, "# %s: %s" % (CMDNAME, parser.name()))
            
            # Output the magic word Autogenerated so we can tell if we can overwrite this
            built_command_output.insert(1, "# Autogenerated from man pages")
            for line in built_command_output:
                output_file.write(line)
                output_file.write('\n')
            output_file.write('\n')
            add_diagnostic(manpage_path + ' parsed successfully')
            if output_file != sys.stdout:
                output_file.close()
        else:
            parser_names =  ', '.join(p.name() for p in parsersToTry)
            #add_diagnostic('%s contains no options or is unparsable' % manpage_path, BRIEF_VERBOSE)
            add_diagnostic('%s contains no options or is unparsable (tried parser %s)' % (manpage_path, parser_names), BRIEF_VERBOSE)
    return success

# Indicates whether the given filename has a presence in one of the yield-to directories
# If so, there's a bespoke completion and we should not generate one
def file_in_yield_directory(filename, yield_to_dirs):
    for yield_dir in yield_to_dirs:
        test_path = os.path.join(yield_dir, filename)
        if os.path.isfile(test_path):
            # Yield to the existing file
            return True
    return False

# Indicates whether we want to skip this command because it already had a non-autogenerated completion
def should_skip_man_page(output_path, filename, yield_to_dirs):
    # No reason to skip if we're writing to stdout
    if WRITE_TO_STDOUT:
        return false
        
    # Check all the yield directories
    for yield_dir in yield_to_dirs:
        test_path = os.path.join(yield_dir, filename)
        if os.path.isfile(test_path):
            # Yield to the existing file
            return true
        
    # See if there's a hand-written file already
    if not file_missing_or_overwritable(output_path):
        return true
    
    # We made it through, so don't skip
    return false
            
    

def parse_and_output_man_pages(paths, output_directory, yield_to_dirs, show_progress):
    global diagnostic_indent, CMDNAME
    paths.sort()
    total_count = len(paths)
    successful_count, index = 0, 0
    padding_len = len(str(total_count))
    last_progress_string_length = 0
    if show_progress and not WRITE_TO_STDOUT:
        print("Parsing man pages and writing completions to {0}".format(output_directory))
    for manpage_path in paths:
        index += 1
        
        # Get the "base" command, e.g. gcc.1.gz -> gcc
        man_file_name = os.path.basename(manpage_path)
        CMDNAME = man_file_name.split('.', 1)[0]
        output_file_name = CMDNAME + '.fish'
        
        # Show progress if we're doing that
        if show_progress:
            progress_str = '  {0} / {1} : {2}'.format((str(index).rjust(padding_len)), total_count, man_file_name)
            # Pad on the right with spaces so we overwrite whatever we wrote last time
            padded_progress_str = progress_str.ljust(last_progress_string_length)
            last_progress_string_length = len(progress_str)
            sys.stdout.write("\r{0}\r".format(padded_progress_str))
            sys.stdout.flush()
                        
        # Maybe we want to skip this item
        skip = False
        if not WRITE_TO_STDOUT:
            # Compute the path that we would write to
            output_path = os.path.join(output_directory, output_file_name)

            if file_in_yield_directory(output_file_name, yield_to_dirs):
                # We're duplicating a bespoke completion - delete any existing completion
                skip = True
                cleanup_autogenerated_file(output_path)
            elif not file_missing_or_overwritable(output_path):
                # Don't overwrite a user-created completion
                skip = True

        # Now skip if requested
        if skip:
            continue
        
        try:
            if parse_manpage_at_path(manpage_path, yield_to_dirs, output_directory):
                successful_count += 1
        except IOError:
            diagnostic_indent = 0
            add_diagnostic('Cannot open ' + manpage_path)
        except (KeyboardInterrupt, SystemExit):
            raise
        except:
            add_diagnostic('Error parsing %s: %s' % (manpage_path, sys.exc_info()[0]), BRIEF_VERBOSE)
            flush_diagnostics(sys.stderr)
            traceback.print_exc(file=sys.stderr)
        flush_diagnostics(sys.stderr)
    print("") #Newline after loop
    add_diagnostic("Successfully parsed %d / %d pages" % (successful_count, total_count), BRIEF_VERBOSE)
    flush_diagnostics(sys.stderr)

def get_paths_from_manpath():
    # Return all the paths to man(1) files in the manpath
    import subprocess, os
    proc = subprocess.Popen(['manpath'], stdout=subprocess.PIPE)
    manpath, err_data = proc.communicate()
    parent_paths = manpath.decode().strip().split(':')
    if not parent_paths:
        sys.stderr.write("Unable to get the manpath (tried manpath)\n")
        sys.exit(-1)
    result = []
    for parent_path in parent_paths:
        directory_path = os.path.join(parent_path, 'man1')
        try:
            names = os.listdir(directory_path)
        except OSError as e:
            names = []
        names.sort()
        for name in names:
            result.append(os.path.join(directory_path, name))
    return result
            
        

def usage(script_name):
    print("Usage: {0} [-v, --verbose] [-s, --stdout] [-d, --directory] [-p, --progress] files...".format(script_name))
    print("""Command options are:
     -h, --help\t\tShow this help message
     -v, --verbose [0, 1, 2]\tShow debugging output to stderr. Larger is more verbose.
     -s, --stdout\tWrite all completions to stdout (trumps the --directory option)
     -d, --directory [dir]\tWrite all completions to the given directory, instead of to ~/.config/fish/completions
     -y, --yield-to [dir]\tSkip completions that are already present in the given directory
     -m, --manpath\tProcess all man1 files available in the manpath (as determined by manpath)
     -p, --progress\tShow progress
    """)

if __name__ == "__main__":
    script_name = sys.argv[0]
    try:
        opts, file_paths = getopt.gnu_getopt(sys.argv[1:], 'v:sd:hmpy:z', ['verbose=', 'stdout', 'directory=', 'help', 'manpath', 'progress', 'yield-to='])
    except getopt.GetoptError as err:
        print(err.msg) # will print something like "option -a not recognized"
        usage(script_name)
        sys.exit(2)
    
    # If a completion already exists in one of the yield-to directories, then don't overwrite it
    # And even delete an existing autogenerated one
    yield_to_dirs = []
    
    use_manpath, show_progress, custom_dir = False, False, False
    output_directory = ''
    for opt, value in opts:
        if opt in ('-v', '--verbose'):
            VERBOSITY = int(value)
        elif opt in ('-s', '--stdout'):
            WRITE_TO_STDOUT = True
        elif opt in ('-d', '--directory'):
            output_directory = value
        elif opt in ('-h', '--help'):
            usage(script_name)
            sys.exit(0)
        elif opt in ('-m', '--manpath'):
            use_manpath = True
        elif opt in ('-p', '--progress'):
            show_progress = True
        elif opt in ('-y', '--yield-to'):
            yield_to_dirs.append(value)
            if not os.path.isdir(value):
                sys.stderr.write("Warning: yield-to directory does not exist: '{0}'\n".format(value))
        elif opt in ('-z'):
            DEROFF_ONLY = True
        else:
            assert False, "unhandled option"
    
    if use_manpath:
        # Fetch all man1 files from the manpath
        file_paths.extend(get_paths_from_manpath())
    
    if not file_paths:
        print("No paths specified")
        sys.exit(0)
    
    if not WRITE_TO_STDOUT and not output_directory:
        # Default to ~/.config/fish/completions/
        # Create it if it doesn't exist
        output_directory = os.path.expanduser('~/.config/fish/completions/')
        try:
            os.makedirs(output_directory)
        except OSError as e:
            if e.errno != errno.EEXIST:
                raise
        
    if True:
        parse_and_output_man_pages(file_paths, output_directory, yield_to_dirs, show_progress)
    else:
        # Profiling code
        import cProfile, pstats
        cProfile.run('parse_and_output_man_pages(file_paths, output_directory, yield_to_dirs, show_progress)', 'fooprof')
        p = pstats.Stats('fooprof')
        p.sort_stats('cumulative').print_stats(100)
    
    # Here we can write out all the parser infos
    if False:
        for name in PARSER_INFO:
            print('Parser ' + name + ':')
            print('\t' + ', '.join(PARSER_INFO[name]))
            print('')