summaryrefslogtreecommitdiff
path: root/update-error-constants.py
blob: 8a8368567aa5482bd8a8126c5e4d68b7fe65e027 (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
#!/usr/bin/env python

from __future__ import print_function, absolute_import

import sys, os, os.path, re, codecs

BUILD_SOURCE_FILE = os.path.join("src", "lxml", "xmlerror.pxi")
BUILD_DEF_FILE    = os.path.join("src", "lxml", "includes", "xmlerror.pxd")

if len(sys.argv) < 2 or sys.argv[1].lower() in ('-h', '--help'):
    print("This script generates the constants in file %s" % BUILD_SOURCE_FILE)
    print("Call as")
    print(sys.argv[0], "/path/to/libxml2-doc-dir")
    sys.exit(len(sys.argv) > 1)

HTML_DIR = os.path.join(sys.argv[1], 'html')
os.stat(HTML_DIR) # raise an error if we can't find it

sys.path.insert(0, 'src')
from lxml import etree

# map enum name to Python variable name and alignment for constant name
ENUM_MAP = {
    'xmlErrorLevel'       : ('__ERROR_LEVELS',  'XML_ERR_'),
    'xmlErrorDomain'      : ('__ERROR_DOMAINS', 'XML_FROM_'),
    'xmlParserErrors'     : ('__PARSER_ERROR_TYPES',   'XML_'),
#    'xmlXPathError'       : ('__XPATH_ERROR_TYPES',   ''),
#    'xmlSchemaValidError' : ('__XMLSCHEMA_ERROR_TYPES',   'XML_'),
    'xmlRelaxNGValidErr'  : ('__RELAXNG_ERROR_TYPES',   'XML_'),
    }

ENUM_ORDER = (
    'xmlErrorLevel',
    'xmlErrorDomain',
    'xmlParserErrors',
#    'xmlXPathError',
#    'xmlSchemaValidError',
    'xmlRelaxNGValidErr')

COMMENT = """
# This section is generated by the script '%s'.

""" % os.path.basename(sys.argv[0])

def split(lines):
    lines = iter(lines)
    pre = []
    for line in lines:
        pre.append(line)
        if line.startswith('#') and "BEGIN: GENERATED CONSTANTS" in line:
            break
    pre.append('')
    for line in lines:
        if line.startswith('#') and "END: GENERATED CONSTANTS" in line:
            break
    post = ['', line]
    post.extend(lines)
    post.append('')
    return pre, post

def regenerate_file(filename, result):
    # read .pxi source file
    f = codecs.open(filename, 'r', encoding="utf-8")
    pre, post = split(f)
    f.close()

    # write .pxi source file
    f = codecs.open(filename, 'w', encoding="utf-8")
    f.write(''.join(pre))
    f.write(COMMENT)
    f.write('\n'.join(result))
    f.write(''.join(post))
    f.close()

collect_text = etree.XPath("string()")
find_enums = etree.XPath(
    "//html:pre[@class = 'programlisting' and contains(text(), 'Enum')]",
    namespaces = {'html' : 'http://www.w3.org/1999/xhtml'})

def parse_enums(html_dir, html_filename, enum_dict):
    PARSE_ENUM_NAME  = re.compile(r'\s*enum\s+(\w+)\s*{', re.I).match
    PARSE_ENUM_VALUE = re.compile(r'\s*=\s+([0-9]+)\s*(?::\s*(.*))?').match
    tree = etree.parse(os.path.join(html_dir, html_filename))
    enums = find_enums(tree)
    for enum in enums:
        enum_name = PARSE_ENUM_NAME(collect_text(enum))
        if not enum_name:
            continue
        enum_name = enum_name.group(1)
        if enum_name not in ENUM_MAP:
            continue
        print("Found enum", enum_name)
        entries = []
        for child in enum:
            name = child.text
            match = PARSE_ENUM_VALUE(child.tail)
            if not match:
                print("Ignoring enum %s (failed to parse field '%s')" % (
                        enum_name, name))
                break
            value, descr = match.groups()
            entries.append((name, int(value), descr))
        else:
            enum_dict[enum_name] = entries
    return enum_dict

enum_dict = {}
parse_enums(HTML_DIR, 'libxml-xmlerror.html',   enum_dict)
#parse_enums(HTML_DIR, 'libxml-xpath.html',      enum_dict)
#parse_enums(HTML_DIR, 'libxml-xmlschemas.html', enum_dict)
parse_enums(HTML_DIR, 'libxml-relaxng.html',    enum_dict)

# regenerate source files
pxi_result = []
append_pxi = pxi_result.append
pxd_result = []
append_pxd = pxd_result.append

append_pxd('cdef extern from "libxml/xmlerror.h":')

ctypedef_indent = ' '*4
constant_indent = ctypedef_indent*2

for enum_name in ENUM_ORDER:
    constants = enum_dict[enum_name]
    pxi_name, prefix = ENUM_MAP[enum_name]

    append_pxd(ctypedef_indent + 'ctypedef enum %s:' % enum_name)
    append_pxi('cdef object %s = """\\' % pxi_name)

    prefix_len = len(prefix)
    length = 2  # each string ends with '\n\0'
    for name, val, descr in constants:
        if descr and descr != str(val):
            line = '%-50s = %7d # %s' % (name, val, descr)
        else:
            line = '%-50s = %7d' % (name, val)
        append_pxd(constant_indent + line)

        if name[:prefix_len] == prefix and len(name) > prefix_len:
            name = name[prefix_len:]
        line = '%s=%d' % (name, val)
        append_pxi(line)
        length += len(line) + 2  # + '\n\0'

    append_pxd('')
    append_pxi('"""')
    append_pxi('')

# write source files
print("Updating file %s" % BUILD_SOURCE_FILE)
regenerate_file(BUILD_SOURCE_FILE, pxi_result)

print("Updating file %s" % BUILD_DEF_FILE)
regenerate_file(BUILD_DEF_FILE,    pxd_result)

print("Done")