Skip to content

Commit 13de09d

Browse files
author
martin.v.loewis
committed
Patch #1359618: Speed-up charmap encoder.
git-svn-id: http://svn.python.org/projects/python/trunk@46648 6015fed2-1504-0410-9fe1-9d1591cc4771
1 parent fb515b3 commit 13de09d

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

51 files changed

+510
-11516
lines changed

Include/unicodeobject.h

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -650,6 +650,11 @@ PyAPI_FUNC(PyObject*) PyUnicode_AsEncodedString(
650650
const char *errors /* error handling */
651651
);
652652

653+
PyAPI_FUNC(PyObject*) PyUnicode_BuildEncodingMap(
654+
PyObject* string /* 256 character map */
655+
);
656+
657+
653658
/* --- UTF-7 Codecs ------------------------------------------------------- */
654659

655660
PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7(

Lib/encodings/cp037.py

Lines changed: 4 additions & 261 deletions
Original file line numberDiff line numberDiff line change
@@ -9,14 +9,14 @@
99
class Codec(codecs.Codec):
1010

1111
def encode(self,input,errors='strict'):
12-
return codecs.charmap_encode(input,errors,encoding_map)
12+
return codecs.charmap_encode(input,errors,encoding_table)
1313

1414
def decode(self,input,errors='strict'):
1515
return codecs.charmap_decode(input,errors,decoding_table)
1616

1717
class IncrementalEncoder(codecs.IncrementalEncoder):
1818
def encode(self, input, final=False):
19-
return codecs.charmap_encode(input,self.errors,encoding_map)[0]
19+
return codecs.charmap_encode(input,self.errors,encoding_table)[0]
2020

2121
class IncrementalDecoder(codecs.IncrementalDecoder):
2222
def decode(self, input, final=False):
@@ -303,263 +303,6 @@ def getregentry():
303303
u'\x9f' # 0xFF -> CONTROL
304304
)
305305

306-
### Encoding Map
306+
### Encoding table
307+
encoding_table=codecs.charmap_build(decoding_table)
307308

308-
encoding_map = {
309-
0x0000: 0x00, # NULL
310-
0x0001: 0x01, # START OF HEADING
311-
0x0002: 0x02, # START OF TEXT
312-
0x0003: 0x03, # END OF TEXT
313-
0x0004: 0x37, # END OF TRANSMISSION
314-
0x0005: 0x2D, # ENQUIRY
315-
0x0006: 0x2E, # ACKNOWLEDGE
316-
0x0007: 0x2F, # BELL
317-
0x0008: 0x16, # BACKSPACE
318-
0x0009: 0x05, # HORIZONTAL TABULATION
319-
0x000A: 0x25, # LINE FEED
320-
0x000B: 0x0B, # VERTICAL TABULATION
321-
0x000C: 0x0C, # FORM FEED
322-
0x000D: 0x0D, # CARRIAGE RETURN
323-
0x000E: 0x0E, # SHIFT OUT
324-
0x000F: 0x0F, # SHIFT IN
325-
0x0010: 0x10, # DATA LINK ESCAPE
326-
0x0011: 0x11, # DEVICE CONTROL ONE
327-
0x0012: 0x12, # DEVICE CONTROL TWO
328-
0x0013: 0x13, # DEVICE CONTROL THREE
329-
0x0014: 0x3C, # DEVICE CONTROL FOUR
330-
0x0015: 0x3D, # NEGATIVE ACKNOWLEDGE
331-
0x0016: 0x32, # SYNCHRONOUS IDLE
332-
0x0017: 0x26, # END OF TRANSMISSION BLOCK
333-
0x0018: 0x18, # CANCEL
334-
0x0019: 0x19, # END OF MEDIUM
335-
0x001A: 0x3F, # SUBSTITUTE
336-
0x001B: 0x27, # ESCAPE
337-
0x001C: 0x1C, # FILE SEPARATOR
338-
0x001D: 0x1D, # GROUP SEPARATOR
339-
0x001E: 0x1E, # RECORD SEPARATOR
340-
0x001F: 0x1F, # UNIT SEPARATOR
341-
0x0020: 0x40, # SPACE
342-
0x0021: 0x5A, # EXCLAMATION MARK
343-
0x0022: 0x7F, # QUOTATION MARK
344-
0x0023: 0x7B, # NUMBER SIGN
345-
0x0024: 0x5B, # DOLLAR SIGN
346-
0x0025: 0x6C, # PERCENT SIGN
347-
0x0026: 0x50, # AMPERSAND
348-
0x0027: 0x7D, # APOSTROPHE
349-
0x0028: 0x4D, # LEFT PARENTHESIS
350-
0x0029: 0x5D, # RIGHT PARENTHESIS
351-
0x002A: 0x5C, # ASTERISK
352-
0x002B: 0x4E, # PLUS SIGN
353-
0x002C: 0x6B, # COMMA
354-
0x002D: 0x60, # HYPHEN-MINUS
355-
0x002E: 0x4B, # FULL STOP
356-
0x002F: 0x61, # SOLIDUS
357-
0x0030: 0xF0, # DIGIT ZERO
358-
0x0031: 0xF1, # DIGIT ONE
359-
0x0032: 0xF2, # DIGIT TWO
360-
0x0033: 0xF3, # DIGIT THREE
361-
0x0034: 0xF4, # DIGIT FOUR
362-
0x0035: 0xF5, # DIGIT FIVE
363-
0x0036: 0xF6, # DIGIT SIX
364-
0x0037: 0xF7, # DIGIT SEVEN
365-
0x0038: 0xF8, # DIGIT EIGHT
366-
0x0039: 0xF9, # DIGIT NINE
367-
0x003A: 0x7A, # COLON
368-
0x003B: 0x5E, # SEMICOLON
369-
0x003C: 0x4C, # LESS-THAN SIGN
370-
0x003D: 0x7E, # EQUALS SIGN
371-
0x003E: 0x6E, # GREATER-THAN SIGN
372-
0x003F: 0x6F, # QUESTION MARK
373-
0x0040: 0x7C, # COMMERCIAL AT
374-
0x0041: 0xC1, # LATIN CAPITAL LETTER A
375-
0x0042: 0xC2, # LATIN CAPITAL LETTER B
376-
0x0043: 0xC3, # LATIN CAPITAL LETTER C
377-
0x0044: 0xC4, # LATIN CAPITAL LETTER D
378-
0x0045: 0xC5, # LATIN CAPITAL LETTER E
379-
0x0046: 0xC6, # LATIN CAPITAL LETTER F
380-
0x0047: 0xC7, # LATIN CAPITAL LETTER G
381-
0x0048: 0xC8, # LATIN CAPITAL LETTER H
382-
0x0049: 0xC9, # LATIN CAPITAL LETTER I
383-
0x004A: 0xD1, # LATIN CAPITAL LETTER J
384-
0x004B: 0xD2, # LATIN CAPITAL LETTER K
385-
0x004C: 0xD3, # LATIN CAPITAL LETTER L
386-
0x004D: 0xD4, # LATIN CAPITAL LETTER M
387-
0x004E: 0xD5, # LATIN CAPITAL LETTER N
388-
0x004F: 0xD6, # LATIN CAPITAL LETTER O
389-
0x0050: 0xD7, # LATIN CAPITAL LETTER P
390-
0x0051: 0xD8, # LATIN CAPITAL LETTER Q
391-
0x0052: 0xD9, # LATIN CAPITAL LETTER R
392-
0x0053: 0xE2, # LATIN CAPITAL LETTER S
393-
0x0054: 0xE3, # LATIN CAPITAL LETTER T
394-
0x0055: 0xE4, # LATIN CAPITAL LETTER U
395-
0x0056: 0xE5, # LATIN CAPITAL LETTER V
396-
0x0057: 0xE6, # LATIN CAPITAL LETTER W
397-
0x0058: 0xE7, # LATIN CAPITAL LETTER X
398-
0x0059: 0xE8, # LATIN CAPITAL LETTER Y
399-
0x005A: 0xE9, # LATIN CAPITAL LETTER Z
400-
0x005B: 0xBA, # LEFT SQUARE BRACKET
401-
0x005C: 0xE0, # REVERSE SOLIDUS
402-
0x005D: 0xBB, # RIGHT SQUARE BRACKET
403-
0x005E: 0xB0, # CIRCUMFLEX ACCENT
404-
0x005F: 0x6D, # LOW LINE
405-
0x0060: 0x79, # GRAVE ACCENT
406-
0x0061: 0x81, # LATIN SMALL LETTER A
407-
0x0062: 0x82, # LATIN SMALL LETTER B
408-
0x0063: 0x83, # LATIN SMALL LETTER C
409-
0x0064: 0x84, # LATIN SMALL LETTER D
410-
0x0065: 0x85, # LATIN SMALL LETTER E
411-
0x0066: 0x86, # LATIN SMALL LETTER F
412-
0x0067: 0x87, # LATIN SMALL LETTER G
413-
0x0068: 0x88, # LATIN SMALL LETTER H
414-
0x0069: 0x89, # LATIN SMALL LETTER I
415-
0x006A: 0x91, # LATIN SMALL LETTER J
416-
0x006B: 0x92, # LATIN SMALL LETTER K
417-
0x006C: 0x93, # LATIN SMALL LETTER L
418-
0x006D: 0x94, # LATIN SMALL LETTER M
419-
0x006E: 0x95, # LATIN SMALL LETTER N
420-
0x006F: 0x96, # LATIN SMALL LETTER O
421-
0x0070: 0x97, # LATIN SMALL LETTER P
422-
0x0071: 0x98, # LATIN SMALL LETTER Q
423-
0x0072: 0x99, # LATIN SMALL LETTER R
424-
0x0073: 0xA2, # LATIN SMALL LETTER S
425-
0x0074: 0xA3, # LATIN SMALL LETTER T
426-
0x0075: 0xA4, # LATIN SMALL LETTER U
427-
0x0076: 0xA5, # LATIN SMALL LETTER V
428-
0x0077: 0xA6, # LATIN SMALL LETTER W
429-
0x0078: 0xA7, # LATIN SMALL LETTER X
430-
0x0079: 0xA8, # LATIN SMALL LETTER Y
431-
0x007A: 0xA9, # LATIN SMALL LETTER Z
432-
0x007B: 0xC0, # LEFT CURLY BRACKET
433-
0x007C: 0x4F, # VERTICAL LINE
434-
0x007D: 0xD0, # RIGHT CURLY BRACKET
435-
0x007E: 0xA1, # TILDE
436-
0x007F: 0x07, # DELETE
437-
0x0080: 0x20, # CONTROL
438-
0x0081: 0x21, # CONTROL
439-
0x0082: 0x22, # CONTROL
440-
0x0083: 0x23, # CONTROL
441-
0x0084: 0x24, # CONTROL
442-
0x0085: 0x15, # CONTROL
443-
0x0086: 0x06, # CONTROL
444-
0x0087: 0x17, # CONTROL
445-
0x0088: 0x28, # CONTROL
446-
0x0089: 0x29, # CONTROL
447-
0x008A: 0x2A, # CONTROL
448-
0x008B: 0x2B, # CONTROL
449-
0x008C: 0x2C, # CONTROL
450-
0x008D: 0x09, # CONTROL
451-
0x008E: 0x0A, # CONTROL
452-
0x008F: 0x1B, # CONTROL
453-
0x0090: 0x30, # CONTROL
454-
0x0091: 0x31, # CONTROL
455-
0x0092: 0x1A, # CONTROL
456-
0x0093: 0x33, # CONTROL
457-
0x0094: 0x34, # CONTROL
458-
0x0095: 0x35, # CONTROL
459-
0x0096: 0x36, # CONTROL
460-
0x0097: 0x08, # CONTROL
461-
0x0098: 0x38, # CONTROL
462-
0x0099: 0x39, # CONTROL
463-
0x009A: 0x3A, # CONTROL
464-
0x009B: 0x3B, # CONTROL
465-
0x009C: 0x04, # CONTROL
466-
0x009D: 0x14, # CONTROL
467-
0x009E: 0x3E, # CONTROL
468-
0x009F: 0xFF, # CONTROL
469-
0x00A0: 0x41, # NO-BREAK SPACE
470-
0x00A1: 0xAA, # INVERTED EXCLAMATION MARK
471-
0x00A2: 0x4A, # CENT SIGN
472-
0x00A3: 0xB1, # POUND SIGN
473-
0x00A4: 0x9F, # CURRENCY SIGN
474-
0x00A5: 0xB2, # YEN SIGN
475-
0x00A6: 0x6A, # BROKEN BAR
476-
0x00A7: 0xB5, # SECTION SIGN
477-
0x00A8: 0xBD, # DIAERESIS
478-
0x00A9: 0xB4, # COPYRIGHT SIGN
479-
0x00AA: 0x9A, # FEMININE ORDINAL INDICATOR
480-
0x00AB: 0x8A, # LEFT-POINTING DOUBLE ANGLE QUOTATION MARK
481-
0x00AC: 0x5F, # NOT SIGN
482-
0x00AD: 0xCA, # SOFT HYPHEN
483-
0x00AE: 0xAF, # REGISTERED SIGN
484-
0x00AF: 0xBC, # MACRON
485-
0x00B0: 0x90, # DEGREE SIGN
486-
0x00B1: 0x8F, # PLUS-MINUS SIGN
487-
0x00B2: 0xEA, # SUPERSCRIPT TWO
488-
0x00B3: 0xFA, # SUPERSCRIPT THREE
489-
0x00B4: 0xBE, # ACUTE ACCENT
490-
0x00B5: 0xA0, # MICRO SIGN
491-
0x00B6: 0xB6, # PILCROW SIGN
492-
0x00B7: 0xB3, # MIDDLE DOT
493-
0x00B8: 0x9D, # CEDILLA
494-
0x00B9: 0xDA, # SUPERSCRIPT ONE
495-
0x00BA: 0x9B, # MASCULINE ORDINAL INDICATOR
496-
0x00BB: 0x8B, # RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK
497-
0x00BC: 0xB7, # VULGAR FRACTION ONE QUARTER
498-
0x00BD: 0xB8, # VULGAR FRACTION ONE HALF
499-
0x00BE: 0xB9, # VULGAR FRACTION THREE QUARTERS
500-
0x00BF: 0xAB, # INVERTED QUESTION MARK
501-
0x00C0: 0x64, # LATIN CAPITAL LETTER A WITH GRAVE
502-
0x00C1: 0x65, # LATIN CAPITAL LETTER A WITH ACUTE
503-
0x00C2: 0x62, # LATIN CAPITAL LETTER A WITH CIRCUMFLEX
504-
0x00C3: 0x66, # LATIN CAPITAL LETTER A WITH TILDE
505-
0x00C4: 0x63, # LATIN CAPITAL LETTER A WITH DIAERESIS
506-
0x00C5: 0x67, # LATIN CAPITAL LETTER A WITH RING ABOVE
507-
0x00C6: 0x9E, # LATIN CAPITAL LIGATURE AE
508-
0x00C7: 0x68, # LATIN CAPITAL LETTER C WITH CEDILLA
509-
0x00C8: 0x74, # LATIN CAPITAL LETTER E WITH GRAVE
510-
0x00C9: 0x71, # LATIN CAPITAL LETTER E WITH ACUTE
511-
0x00CA: 0x72, # LATIN CAPITAL LETTER E WITH CIRCUMFLEX
512-
0x00CB: 0x73, # LATIN CAPITAL LETTER E WITH DIAERESIS
513-
0x00CC: 0x78, # LATIN CAPITAL LETTER I WITH GRAVE
514-
0x00CD: 0x75, # LATIN CAPITAL LETTER I WITH ACUTE
515-
0x00CE: 0x76, # LATIN CAPITAL LETTER I WITH CIRCUMFLEX
516-
0x00CF: 0x77, # LATIN CAPITAL LETTER I WITH DIAERESIS
517-
0x00D0: 0xAC, # LATIN CAPITAL LETTER ETH (ICELANDIC)
518-
0x00D1: 0x69, # LATIN CAPITAL LETTER N WITH TILDE
519-
0x00D2: 0xED, # LATIN CAPITAL LETTER O WITH GRAVE
520-
0x00D3: 0xEE, # LATIN CAPITAL LETTER O WITH ACUTE
521-
0x00D4: 0xEB, # LATIN CAPITAL LETTER O WITH CIRCUMFLEX
522-
0x00D5: 0xEF, # LATIN CAPITAL LETTER O WITH TILDE
523-
0x00D6: 0xEC, # LATIN CAPITAL LETTER O WITH DIAERESIS
524-
0x00D7: 0xBF, # MULTIPLICATION SIGN
525-
0x00D8: 0x80, # LATIN CAPITAL LETTER O WITH STROKE
526-
0x00D9: 0xFD, # LATIN CAPITAL LETTER U WITH GRAVE
527-
0x00DA: 0xFE, # LATIN CAPITAL LETTER U WITH ACUTE
528-
0x00DB: 0xFB, # LATIN CAPITAL LETTER U WITH CIRCUMFLEX
529-
0x00DC: 0xFC, # LATIN CAPITAL LETTER U WITH DIAERESIS
530-
0x00DD: 0xAD, # LATIN CAPITAL LETTER Y WITH ACUTE
531-
0x00DE: 0xAE, # LATIN CAPITAL LETTER THORN (ICELANDIC)
532-
0x00DF: 0x59, # LATIN SMALL LETTER SHARP S (GERMAN)
533-
0x00E0: 0x44, # LATIN SMALL LETTER A WITH GRAVE
534-
0x00E1: 0x45, # LATIN SMALL LETTER A WITH ACUTE
535-
0x00E2: 0x42, # LATIN SMALL LETTER A WITH CIRCUMFLEX
536-
0x00E3: 0x46, # LATIN SMALL LETTER A WITH TILDE
537-
0x00E4: 0x43, # LATIN SMALL LETTER A WITH DIAERESIS
538-
0x00E5: 0x47, # LATIN SMALL LETTER A WITH RING ABOVE
539-
0x00E6: 0x9C, # LATIN SMALL LIGATURE AE
540-
0x00E7: 0x48, # LATIN SMALL LETTER C WITH CEDILLA
541-
0x00E8: 0x54, # LATIN SMALL LETTER E WITH GRAVE
542-
0x00E9: 0x51, # LATIN SMALL LETTER E WITH ACUTE
543-
0x00EA: 0x52, # LATIN SMALL LETTER E WITH CIRCUMFLEX
544-
0x00EB: 0x53, # LATIN SMALL LETTER E WITH DIAERESIS
545-
0x00EC: 0x58, # LATIN SMALL LETTER I WITH GRAVE
546-
0x00ED: 0x55, # LATIN SMALL LETTER I WITH ACUTE
547-
0x00EE: 0x56, # LATIN SMALL LETTER I WITH CIRCUMFLEX
548-
0x00EF: 0x57, # LATIN SMALL LETTER I WITH DIAERESIS
549-
0x00F0: 0x8C, # LATIN SMALL LETTER ETH (ICELANDIC)
550-
0x00F1: 0x49, # LATIN SMALL LETTER N WITH TILDE
551-
0x00F2: 0xCD, # LATIN SMALL LETTER O WITH GRAVE
552-
0x00F3: 0xCE, # LATIN SMALL LETTER O WITH ACUTE
553-
0x00F4: 0xCB, # LATIN SMALL LETTER O WITH CIRCUMFLEX
554-
0x00F5: 0xCF, # LATIN SMALL LETTER O WITH TILDE
555-
0x00F6: 0xCC, # LATIN SMALL LETTER O WITH DIAERESIS
556-
0x00F7: 0xE1, # DIVISION SIGN
557-
0x00F8: 0x70, # LATIN SMALL LETTER O WITH STROKE
558-
0x00F9: 0xDD, # LATIN SMALL LETTER U WITH GRAVE
559-
0x00FA: 0xDE, # LATIN SMALL LETTER U WITH ACUTE
560-
0x00FB: 0xDB, # LATIN SMALL LETTER U WITH CIRCUMFLEX
561-
0x00FC: 0xDC, # LATIN SMALL LETTER U WITH DIAERESIS
562-
0x00FD: 0x8D, # LATIN SMALL LETTER Y WITH ACUTE
563-
0x00FE: 0x8E, # LATIN SMALL LETTER THORN (ICELANDIC)
564-
0x00FF: 0xDF, # LATIN SMALL LETTER Y WITH DIAERESIS
565-
}

0 commit comments

Comments
 (0)