diff options
Diffstat (limited to 'examples/corelib/serialization/cbordump/cbortag.py')
| -rwxr-xr-x | examples/corelib/serialization/cbordump/cbortag.py | 188 |
1 files changed, 188 insertions, 0 deletions
diff --git a/examples/corelib/serialization/cbordump/cbortag.py b/examples/corelib/serialization/cbordump/cbortag.py new file mode 100755 index 0000000000..26a0f969e4 --- /dev/null +++ b/examples/corelib/serialization/cbordump/cbortag.py @@ -0,0 +1,188 @@ +#!/bin/env python3 +# Copyright (C) 2023 The Qt Company Ltd. +# SPDX-License-Identifier: LicenseRef-Qt-Commercial OR BSD-3-Clause +"""Digest cbor-tags.xml file into code for insertion into main.cpp + +See main.cpp's comment on how to regenerate its GENERATED CODE. +See ./cbortag.py --help for further details on how to invoke. +You can import this is a module without invoking the script. +""" + +def firstChild(parent, tag): + """Return parent's first child element with the given tag.""" + return next(node for node in parent.childNodes + if node.nodeType == parent.ELEMENT_NODE and node.nodeName == tag) + +def nodeAttrIs(node, attr, seek): + """Checks whether the node has a given value for an attribute + + Takes the node to check, the name of the attribute and the value + to check against. Returns true if the node does have that value + for the named attribute.""" + if node.nodeType != node.ELEMENT_NODE: + return False + if node.attributes is None or attr not in node.attributes: + return False + return node.attributes[attr].value == seek + +def getRfcValue(node): + """Extract RFC reference from an <xref type="rfc" ...> element + + Some of these have a reference including section details as the + body of the element, otherwise the data attribute should identify + the RFC. If neither is found, an empty string is returned.""" + if node.childNodes: + return node.childNodes[0].nodeValue # Maybe accumulate several children ? + if node.attributes is None or 'data' not in node.attributes: + return '' + return node.attributes['data'].value + +def readRegistry(filename): + """Handles the XML parsing and returns the relevant parts. + + Single argument is the path to the cbor-tags.xml file; returns a + twople of the title element's text and an interator over the + record nodes. Checks some things are as expected while doing so.""" + from xml.dom.minidom import parse + doc = parse(filename).documentElement + assert nodeAttrIs(doc, 'id', 'cbor-tags') + title = firstChild(doc, 'title').childNodes[0].nodeValue + registry = firstChild(doc, 'registry') + assert nodeAttrIs(registry, 'id', 'tags') + records = (node for node in registry.childNodes if node.nodeName == 'record') + return title, records + +def digest(record): + """Digest a single record from cbor-tags.xml + + If the record is not of interest, returns the twople (None, None). + For records of interest, returns (n, t) where n is the numeric tag + code of the record and t is a text describing it. If the record, + or its semantics field, has an xref child with type="rfc", the RFC + mentioned there is included with the text of the semantics; such a + record is of interest, provided it has a semantics field and no + dash in its value. Records with a value field containing a dash + (indicating a range) are not of interest. Records with a value of + 256 or above are only of interest if they include an RFC.""" + data = {} + for kid in record.childNodes: + if kid.nodeName == 'xref': + if not nodeAttrIs(kid, 'type', 'rfc'): + continue + rfc = getRfcValue(kid) + if rfc: + # Potentially stomping one taken from semantics + data['rfc'] = rfc + elif kid.nodeName == 'semantics': + text = rfc = '' + for part in kid.childNodes: + if part.nodeType == kid.TEXT_NODE: + text += part.nodeValue + elif part.nodeType == kid.ELEMENT_NODE: + if part.nodeName != 'xref' or not nodeAttrIs(part, 'type', 'rfc'): + continue # potentially append content to text + assert not rfc, ('Duplicate RFC ?', rfc, part) + rfc = getRfcValue(part) + if rfc: + if text.endswith('()'): + text = text[:-2].rstrip() + if 'rfc' not in data: + data['rfc'] = rfc + data['semantics'] = ' '.join(text.split()) + elif kid.nodeName == 'value': + data['value'] = kid.childNodes[0].nodeValue + text = data.get('semantics') + if not text or 'value' not in data or '-' in data['value']: + return None, None + value = int(data['value']) + if 'rfc' in data: + rfc = data["rfc"].replace('rfc', 'RFC') + text = f'{text} [{rfc}]' + elif value >= 256: + return None, None + return value, text + +def entries(records): + """Digest each record of interest into a value and text. + + The value and text form the raw material of the tagDescriptions + array in main.cpp; see digest for which records are retained.""" + for record in records: + value, text = digest(record) + if value is not None: + yield value, text + +def marginBound(text, prior, left, right): + """Split up a string literal for tidy display. + + The first parameter, text, is the content of the string literal; + quotes shall be added. It may be split into several fragments, + each quoted, so as to abide by line length constraints. + + The remaining parameters are integers: prior is the text already + present on the line before text is to be added; left is the width + of the left margin for all subsequent lines; and right is the + right margin to stay within, where possible. The returned string + is either a space with the whole quoted text following, to fit on + the line already started to length prior, or a sequence of quoted + strings, each preceded by a newline and indent of width left.""" + if prior + 3 + len(text) < right: # 1 for space, 2 for quotes + return f' "{text}"' + width = right - left - 2 # 2 for the quotes + words = iter(text.split(' ')) + lines, current = [''], [next(words)] + for word in words: + if len(word) + sum(len(w) + 1 for w in current) > width: + line = ' '.join(current) + lines.append(f'"{line}"') + current = ['', word] + else: + current.append(word) + line = ' '.join(current) + lines.append(f'"{line}"') + return ('\n' + ' ' * left).join(lines) + +def main(argv, speak): + """Takes care of driving the process. + + Takes the command-line argument list (whose first entry is the + name of this script) and standard output (or compatible stream of + your choosing) to which to write data. If the --out option is + specified in the arguments, the file it names is used in place of + this output stream.""" + from argparse import ArgumentParser, ArgumentDefaultsHelpFormatter + parser = ArgumentParser( + description='Digest cbor-tags.xml into code to insert in main.cpp', + formatter_class=ArgumentDefaultsHelpFormatter) + parser.add_argument('path', help='path of the cbor-tags.xml file', + default='cbor-tags.xml') + parser.add_argument('--out', help='file to write instead of standard output') + args = parser.parse_args(argv[1:]) + emit = (open(args.out) if args.out else speak).write + + title, records = readRegistry(args.path) + emit(f"""\ +struct CborTagDescription +{{ + QCborTag tag; + const char *description; // with space and parentheses +}}; + +// {title} +static const CborTagDescription tagDescriptions[] = {{ + // from https://www.iana.org/assignments/cbor-tags/cbor-tags.xhtml +""") + + for value, text in sorted(entries(records)): + prior = f' {{ QCborTag({value}),' + body = marginBound(f' ({text})', len(prior), 6, 96) + emit(f"{prior}{body} }},\n") + + emit("""\ + { QCborTag(-1), nullptr } +}; +""") + +if __name__ == '__main__': + import sys + sys.exit(main(sys.argv, sys.stdout)) |