From d5b501c4046fbdc2354607c12f9f8230159e6135 Mon Sep 17 00:00:00 2001 From: Rob Austein Date: Fri, 9 Jul 2021 21:21:02 +0000 Subject: Incorporate project-independent stuff I did for IETF NOC --- .gitmodules | 4 + GNUmakefile | 18 +- Makefile | 4 +- README | 32 +-- extract.py | 124 ---------- filter.json | 19 -- pelicanconf.py | 34 +-- references/convert-and-slurp-attachments.sh | 18 -- references/extract-wiki-content.xsl | 177 --------------- references/generate-json.py | 154 ------------- references/pelicanconf.py | 42 ---- references/publishconf.py | 19 -- references/rpki-wiki-to-markdown.py | 341 ---------------------------- references/schema.sql | 177 --------------- references/trac-wiki-to-markdown.rb | 51 ----- references/trac2down.py | 61 ----- tools | 1 + trac2md.py | 193 ---------------- 18 files changed, 10 insertions(+), 1459 deletions(-) create mode 100644 .gitmodules mode change 100644 => 120000 GNUmakefile mode change 100644 => 120000 Makefile delete mode 100755 extract.py delete mode 100644 filter.json mode change 100644 => 120000 pelicanconf.py delete mode 100755 references/convert-and-slurp-attachments.sh delete mode 100644 references/extract-wiki-content.xsl delete mode 100755 references/generate-json.py delete mode 100644 references/pelicanconf.py delete mode 100644 references/publishconf.py delete mode 100644 references/rpki-wiki-to-markdown.py delete mode 100644 references/schema.sql delete mode 100644 references/trac-wiki-to-markdown.rb delete mode 100644 references/trac2down.py create mode 160000 tools delete mode 100755 trac2md.py diff --git a/.gitmodules b/.gitmodules new file mode 100644 index 0000000..81fb24b --- /dev/null +++ b/.gitmodules @@ -0,0 +1,4 @@ +[submodule "tools"] + path = tools + url = git@git.hactrn.net:sra/trac-wiki-to-markdown.git + branch = trunk diff --git a/GNUmakefile b/GNUmakefile deleted file mode 100644 index 7a13d6d..0000000 --- a/GNUmakefile +++ /dev/null @@ -1,17 +0,0 @@ -all: - ./extract.py - cd pelican; pelican --output website --settings pelicanconf.py --fatal errors content - -fetch: - rsync -aP --delete bikeshed.cryptech.is:/home/trac/db/trac.db bikeshed.cryptech.is:/home/trac/files/attachments . - -clean: - rm -rf wiki pelican - -distclean: clean - rm -rf trac.db attachments - -webfsd: - webfsd -r pelican/website -4 -L - -F -f index.html - -.PHONY: all clean fetch distclean webfsd diff --git a/GNUmakefile b/GNUmakefile new file mode 120000 index 0000000..326927d --- /dev/null +++ b/GNUmakefile @@ -0,0 +1 @@ +tools/GNUmakefile \ No newline at end of file diff --git a/Makefile b/Makefile deleted file mode 100644 index c1aa5dc..0000000 --- a/Makefile +++ /dev/null @@ -1,3 +0,0 @@ -.MAIN: all - -${.TARGETS}: ; @gmake $@ diff --git a/Makefile b/Makefile new file mode 120000 index 0000000..84519ca --- /dev/null +++ b/Makefile @@ -0,0 +1 @@ +tools/Makefile \ No newline at end of file diff --git a/README b/README index 299f3f3..dfdf4fb 100644 --- a/README +++ b/README @@ -1,33 +1,5 @@ -Attempts to convert Permatrac Wiki to Markdown format. - -Last time I had to do something like this it looks like I took the -Trac -> HTML -> html2markdown path. Could do that again. Recipe for -fetching the HTML: - - wget -r --user=sra@hactrn.net --ask-password https://permatrac.noc.ietf.org/wiki/ - -Slightly better recipe: - - wget -m -R '*[?]*' --user=sra@hactrn.net --ask-password https://permatrac.noc.ietf.org/wiki/ - -But either way, `wget` seems to be non-terminating when walking this -site, even if it throws away .999 of what it retrieves. In theory, -adding `-np` would have fixed this; in practice doing so causes it not -to recurse at all. Grrr. - -Scripts from rpki.net/doc/*/*.{sh,py,xsl}, included here under tools/. - -The other approach is converting Trac wiki directly to markdown. -There are hacks for this, in Ruby: - - https://gist.github.com/somebox/619537 - -But there are also Python versions, which are a bit more tractible. -Several of them included in tools/ too. - - -The most promising approach looks to be `tools/trac2md.py`, which is a -bit crude but is written in a relatively sane style so should be fixable. +Convert wiki.cryptech.is to Markdown +See tools/README for general information about the tools Pelican content format: https://docs.getpelican.com/en/latest/content.html diff --git a/extract.py b/extract.py deleted file mode 100755 index 793b502..0000000 --- a/extract.py +++ /dev/null @@ -1,124 +0,0 @@ -#!/usr/bin/env python3 - -import fnmatch -import hashlib -import json -import os -import shutil -import sqlite3 -import sys -import time -import urllib.parse - -import trac2md - -wiki_query = ''' - SELECT - name, - author, - version, - time / 1000000 AS time, - text - FROM wiki - ORDER BY - name, version -''' - -attachment_query = ''' - SELECT - id, - filename, - size, - author, - description, - ipnr, - time / 1000000 AS createdtime - FROM - attachment - WHERE - type = 'wiki' - ORDER BY - filename, time -''' - -def attachment_link(row): - h = lambda whatever: hashlib.sha1(whatever.encode()).hexdigest() - h1 = h(row.id) - h2 = h(row.filename) - fn2 = os.path.splitext(row["filename"])[1] - return \ - os.path.join("attachments", "wiki", h1[:3], h1, h2 + fn2), \ - os.path.join("pelican", "content", urllib.parse.quote(row.id, ""), row.filename) - -class Filter: - - def __init__(self, filename = "filter.json"): - with open(filename) as f: - filter = json.load(f) - if not all(action in "-+" for action, pattern in filter): - sys.exit("Bad action \"{}\" in filter".format(action)) - self.filter = tuple((action == "+", pattern) for action, pattern in filter) - - def __call__(self, name): - for action, pattern in self.filter: - if fnmatch.fnmatch(name, pattern): - return action - return True - -class Row(sqlite3.Row): - - def __getattr__(self, name): - return self[name] - - @property - def isotime(self): - return time.strftime("%Y-%m-%d %H:%M", time.gmtime(self.time)) - -def markdown_header(row, first_published): - if row.name in first_published: - modtime = "Modified: {}\n".format(row.isotime) - else: - modtime = "" - first_published[row.name] = row.isotime - return "Title: {}\nAuthor: {}\nDate: {}\n{}\n".format(row.name, row.author, first_published[row.name], modtime) - -def main(): - - for dn in ("wiki", "pelican"): - shutil.rmtree(dn) - - for dn in ("wiki", "pelican/content"): - os.makedirs(dn) - - os.link("pelicanconf.py", "pelican/pelicanconf.py") - - wiki_to_markdown = trac2md.Trac2Markdown() - - keep = Filter() - - first_published = {} - - db = sqlite3.connect("trac.db") - db.row_factory = Row - - for row in db.execute(wiki_query): - if keep(row.name): - slug = urllib.parse.quote(row.name, "") - #print(slug, row.version) - with open("wiki/{}.trac".format(slug), "w") as f: - f.write(row.text) - md = markdown_header(row, first_published) + wiki_to_markdown(row.text, slug) - with open("pelican/content/{}.md".format(slug), "w") as f: - f.write(md) - - for row in db.execute(attachment_query): - src, dst = attachment_link(row) - #print("{} => {}".format(dst, src)) - if not os.path.isdir(os.path.dirname(dst)): - os.makedirs(os.path.dirname(dst)) - os.link(src, dst) - - db.close() - -if __name__ == "__main__": - main() diff --git a/filter.json b/filter.json deleted file mode 100644 index 708428e..0000000 --- a/filter.json +++ /dev/null @@ -1,19 +0,0 @@ -[ - ["+", "WikiStart"], - ["-", "CamelCase"], - ["-", "EDAToolchainSurvey\""], - ["-", "GitRepositories*"], - ["-", "InterMapTxt"], - ["-", "InterTrac"], - ["-", "InterWiki"], - ["-", "PageTemplates"], - ["-", "PhotoFolder"], - ["-", "RecentChanges"], - ["-", "SandBox"], - ["-", "ScratchPage"], - ["-", "TicketQuery"], - ["-", "TitleIndex"], - ["-", "Trac*"], - ["-", "Wiki*"], - ["+", "*"] -] diff --git a/pelicanconf.py b/pelicanconf.py deleted file mode 100644 index ab8afd2..0000000 --- a/pelicanconf.py +++ /dev/null @@ -1,33 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- # - -AUTHOR = "Cryptech Core Team" -SITENAME = "Cryptech Project" - -PATH = "content" -TIMEZONE = "UTC" -DEFAULT_LANG = "en" - -# Feed generation is usually not desired when developing -SITEURL = "" -RELATIVE_URLS = True -FEED_ALL_ATOM = None -CATEGORY_FEED_ATOM = None -TRANSLATION_FEED_ATOM = None -AUTHOR_FEED_ATOM = None -AUTHOR_FEED_RSS = None - -# Blogroll -LINKS = (("Pelican", "http://getpelican.com/"), - ("Python.org", "http://python.org/"), - ("Jinja2", "http://jinja.pocoo.org/")) -LINKS_WIDGET_NAME = "Links" - -# Social widget. Can't get rid of this with default theme, only change its name. -# Fiddle with themes later -SOCIAL = () -SOCIAL_WIDGET_NAME = "Subscribe" - -DEFAULT_PAGINATION = 10 - -#THEME = "/home/blog/pelican-themes/sundown" diff --git a/pelicanconf.py b/pelicanconf.py new file mode 120000 index 0000000..00f205f --- /dev/null +++ b/pelicanconf.py @@ -0,0 +1 @@ +tools/pelicanconf.py \ No newline at end of file diff --git a/references/convert-and-slurp-attachments.sh b/references/convert-and-slurp-attachments.sh deleted file mode 100755 index ce7f34d..0000000 --- a/references/convert-and-slurp-attachments.sh +++ /dev/null @@ -1,18 +0,0 @@ -#!/bin/sh - - -ls | fgrep -v . | -while read page -do - base="https://trac.rpki.net" - path="/wiki/$(echo $page | sed s=%2F=/=g)" - - # Fetch the Wiki page, extract the useful portion of the HTML, convert that into Markdown - curl "${base}${path}" | - xsltproc --html extract-wiki-content.xsl - | - html2markdown --no-skip-internal-links --reference-links >"$page.md" - - # Fetch a ZIP file containing any attachments, clean up if result is empty or broken - curl "${base}/zip-attachment${path}/" >"$page.zip" - zipinfo "$page.zip" >/dev/null 2>&1 || rm -f "$page.zip" - -done diff --git a/references/extract-wiki-content.xsl b/references/extract-wiki-content.xsl deleted file mode 100644 index e4376e8..0000000 --- a/references/extract-wiki-content.xsl +++ /dev/null @@ -1,177 +0,0 @@ - - - - - - - - - - - - - - - - - - NEW PAGE - - -
- -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - _ - - - / - . - - - - - - - - ​ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - diff --git a/references/generate-json.py b/references/generate-json.py deleted file mode 100755 index b8b1f38..0000000 --- a/references/generate-json.py +++ /dev/null @@ -1,154 +0,0 @@ -#!/usr/bin/env python - -# Generate JSON to import Trac tickets into GitHub issues using the new import API -# described at https://gist.github.com/jonmagic/5282384165e0f86ef105 - -import os -import time -import json -import yaml -import sqlite3 -import hashlib -import argparse -import subprocess - -ticket_query = ''' -SELECT - id, - type, - owner, - reporter, - milestone, - status, - resolution, - summary, - description, - component, - priority, - time / 1000000 AS createdtime, - changetime / 1000000 AS modifiedtime -FROM - ticket -ORDER BY - id -''' - -comment_query = ''' -SELECT - time / 1000000 AS createdtime, - author, - newvalue -FROM - ticket_change -WHERE - ticket = ? -AND - field = 'comment' -AND - newvalue <> '' -ORDER BY - time -''' - -attachment_query = ''' -SELECT - id, - filename, - size, - author, - description, - ipnr, - time / 1000000 AS createdtime -FROM - attachment -WHERE - id = ? -AND - type = 'ticket' -ORDER BY - time, filename -''' - -def isotime(t): - return None if t == 0 else time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime(t)) - -def hashname(whatever): - return hashlib.sha1(unicode(whatever)).hexdigest() - -def ticket_text(ticket): - d = dict(ticket, createdtime = isotime(ticket["createdtime"]), modifiedtime = isotime(ticket["modifiedtime"])) - return u"{description}\n\n" \ - u"_Trac ticket #{id} component {component} priority {priority}, owner {owner}," \ - u" created by {reporter} on {createdtime}, last modified {modifiedtime}_\n".format(**d) - -def comment_text(comment): - d = dict(comment, createdtime = isotime(comment["createdtime"])) - return u"{newvalue}\n\n_Trac comment by {author} on {createdtime}_\n".format(**d) - -def attachment_text(attachment): - h1 = hashname(attachment["id"]) - h2 = hashname(attachment["filename"]) - fn2 = os.path.splitext(attachment["filename"])[1] - fn = os.path.join(gist_url, h1[:3], h1, h2 + fn2) - url = "{}/raw/{}/ticket.{}.{}{}".format(gist_url.rstrip("/"), gist_commit, h1, h2, fn2) - d = dict(attachment, createdtime = isotime(comment["createdtime"]), url = url) - return u"[{filename}]({url}) {description}\n_Trac attachment by {author} on {createdtime}_\n".format(**d) - -def comment_merge(comments, attachments): - result = [] - while comments and attachments: - result.append(comments.pop(0) if comments[0]["created_at"] <= attachments[0]["created_at"] else attachments.pop(0)) - return result + comments + attachments - -parser = argparse.ArgumentParser(formatter_class = argparse.ArgumentDefaultsHelpFormatter) -parser.add_argument("-c", "--config", type = argparse.FileType(), - default = "generate-json.yaml", - help = "YAML config mappings") -args = parser.parse_args() - -cfg = yaml.safe_load(args.config) -assignee_map = cfg["assignees"] -type_map = cfg["type_labels"] -resolution_map = cfg["resolution_labels"] - -gist_url = cfg.get("attachment_gist_url") -if gist_url is not None: - gist_commit = subprocess.check_output(("git", "ls-remote", gist_url, "HEAD")).split()[0] - -db = sqlite3.connect(cfg["database"]) -db.row_factory = sqlite3.Row -ticket_cursor = db.cursor() -comment_cursor = db.cursor() -attachment_cursor = db.cursor() - -if not os.path.isdir(cfg["ticket_directory"]): - os.makedirs(cfg["ticket_directory"]) - -for ticket in ticket_cursor.execute(ticket_query): - comments = comment_merge([dict(created_at = isotime(comment["createdtime"]), body = comment_text(comment)) - for comment in comment_cursor.execute(comment_query, (ticket["id"],))], - [] if gist_url is None else - [dict(created_at = isotime(attachment["createdtime"]), body = attachment_text(attachment)) - for attachment in attachment_cursor.execute(attachment_query, (ticket["id"],))]) - issue = dict( - title = ticket["summary"], - body = ticket_text(ticket), - created_at = isotime(ticket["createdtime"]), - updated_at = isotime(ticket["modifiedtime"])) - if ticket["status"] == "closed": - issue["closed"] = True - issue["closed_at"] = isotime(ticket["modifiedtime"]) - comments.append(dict(created_at = isotime(ticket["modifiedtime"]), - body = "_Closed with resolution {resolution}_\n".format(**ticket))) - if ticket["owner"] in assignee_map: - issue["assignee"] = assignee_map[ticket["owner"]] - labels = [type_map.get(ticket["type"]), resolution_map.get(ticket["resolution"])] - while None in labels: - del labels[labels.index(None)] - if labels: - issue["labels"] = labels - issue = dict(issue = issue) - if comments: - issue["comments"] = comments - with open(os.path.join(cfg["ticket_directory"], "ticket_{:03d}.json".format(ticket["id"])), "wb") as f: - json.dump(issue, f, indent = 4, sort_keys = True, separators=(",", ": ")) diff --git a/references/pelicanconf.py b/references/pelicanconf.py deleted file mode 100644 index a28721d..0000000 --- a/references/pelicanconf.py +++ /dev/null @@ -1,42 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- # -from __future__ import unicode_literals - -AUTHOR = u'Rob Austein' -SITENAME = u'Your Bug Report Will Be Graded' - -# Apparently this is much longer than theme designer expected. Skip it for now. -SITESUBTITLE = u'"I\'m not proud of being a congenital pain in the ass. But I will take money for it."' - -PATH = 'content' -TIMEZONE = 'UTC' -DEFAULT_LANG = u'English' - -# Hack article URLs to match what Blogofile did, to avoid breaking links. - -ARTICLE_URL = '{date:%Y}/{date:%m}/{date:%d}/{slug}/' -ARTICLE_SAVE_AS = '{date:%Y}/{date:%m}/{date:%d}/{slug}/index.html' - -# Feed generation is usually not desired when developing -SITEURL = '' -RELATIVE_URLS = True -FEED_ALL_ATOM = None -CATEGORY_FEED_ATOM = None -TRANSLATION_FEED_ATOM = None -AUTHOR_FEED_ATOM = None -AUTHOR_FEED_RSS = None - -# Blogroll -LINKS = (('Pelican', 'http://getpelican.com/'), - ('Python.org', 'http://python.org/'), - ('Jinja2', 'http://jinja.pocoo.org/')) -LINKS_WIDGET_NAME = "Links" - -# Social widget. Can't get rid of this with default theme, only change its name. -# Fiddle with themes later -SOCIAL = () -SOCIAL_WIDGET_NAME = "Subscribe" - -DEFAULT_PAGINATION = 10 - -THEME = "/home/blog/pelican-themes/sundown" diff --git a/references/publishconf.py b/references/publishconf.py deleted file mode 100644 index f0fb21d..0000000 --- a/references/publishconf.py +++ /dev/null @@ -1,19 +0,0 @@ -#!/usr/local/bin/python2.7 -# -*- coding: utf-8 -*- # -from __future__ import unicode_literals - -# This file is only used if you use `make publish` or -# explicitly specify it as your config file. - -import os -import sys -sys.path.append(os.curdir) -from pelicanconf import * - -SITEURL = 'https://www.hactrn.net/blog' -RELATIVE_URLS = False - -FEED_ALL_ATOM = 'feeds/all.atom.xml' -CATEGORY_FEED_ATOM = 'feeds/{slug}.atom.xml' - -DELETE_OUTPUT_DIRECTORY = True diff --git a/references/rpki-wiki-to-markdown.py b/references/rpki-wiki-to-markdown.py deleted file mode 100644 index dff87e6..0000000 --- a/references/rpki-wiki-to-markdown.py +++ /dev/null @@ -1,341 +0,0 @@ -# Copyright (C) 2016 Parsons Government Services ("PARSONS") -# Portions copyright (C) 2014 Dragon Research Labs ("DRL") -# Portions copyright (C) 2012 Internet Systems Consortium ("ISC") -# -# Permission to use, copy, modify, and distribute this software for any -# purpose with or without fee is hereby granted, provided that the above -# copyright notices and this permission notice appear in all copies. -# -# THE SOFTWARE IS PROVIDED "AS IS" AND PARSONS, DRL, AND ISC DISCLAIM -# ALL WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED -# WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL -# PARSONS, DRL, OR ISC BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR -# CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS -# OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, -# NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION -# WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. - -""" -Trac Wiki -> Markdown converter, hacked from old Trac Wiki -> PDF/flat -text converter. - -Pull HTML pages from a Trac Wiki, feed the useful bits to -html2text to generate Markdown. - -Assumes you're using the TracNav plugin for the Wiki pages, and uses -the same list as the TracNav plugin does to determine the set of pages -to convert. -""" - -# Dependencies, at least on Ubuntu Xenial: -# -# apt-get install python-lxml python-html2text -# -# Be warned that there are many unrelated packages named "html2text", -# installed under various names on various platforms. This one -# happens to be a useful HTML-to-Markdown converter. - -# Most of the work of massaging the HTML is done using XSL transforms, -# because the template-driven style makes that easy. There's probably -# some clever way to use lxml's XPath code to do the same thing in a -# more pythonic way with ElementTrees, but I already had the XSL -# transforms and there's a point of diminishing returns on this sort of -# thing. - -import sys -import os -import argparse -import lxml.etree -import urllib -import urlparse -import subprocess -import zipfile - -# Main program, up front so it doesn't get lost under all the XSL - -def main(): - - base = "https://trac.rpki.net" - - parser = argparse.ArgumentParser(description = __doc__, formatter_class = argparse.ArgumentDefaultsHelpFormatter) - parser.add_argument("-b", "--base_url", - default = base, - help = "base URL for documentation web site") - parser.add_argument("-t", "--toc", - default = base + "/wiki/doc/RPKI/TOC", - help = "table of contents URL") - parser.add_argument("-d", "--directory", - default = ".", - help = "output directory") - parser.add_argument("-p", "--prefix", - default = "/wiki/doc", - help = "page name prefix on wiki") - args = parser.parse_args() - - urls = str(xsl_get_toc(lxml.etree.parse(urllib.urlopen(args.toc)).getroot(), - basename = repr(args.base_url))).splitlines() - - assert all(urlparse.urlparse(url).path.startswith(args.prefix) for url in urls) - - for pagenum, url in enumerate(urls): - path = urlparse.urlparse(url).path - page = xsl_get_page(lxml.etree.parse(urllib.urlopen(url)).getroot(), - basename = repr(args.base_url), - path = repr(path)) - - fn_base = os.path.join(args.directory, "{:02d}{}".format(pagenum, path[len(args.prefix):].replace("/", "."))) - - fn = fn_base + ".zip" - zip_url = urlparse.urljoin(url, "/zip-attachment{}/".format(path)) - urllib.urlretrieve(zip_url, fn) - with zipfile.ZipFile(fn, "r") as z: - if len(z.namelist()) == 0: - os.unlink(fn) - else: - sys.stderr.write("Wrote {}\n".format(fn)) - - for imgnum, img in enumerate(page.xpath("//img | //object | //embed")): - img_url = img.get("data" if img.tag == "object" else "src") - img_url = urlparse.urljoin(url, img_url) - fn = "{}.{:02d}{}".format(fn_base, imgnum, os.path.splitext(img_url)[1]) - urllib.urlretrieve(img_url, fn) - sys.stderr.write("Wrote {}\n".format(fn)) - - html2markdown = subprocess.Popen(("html2markdown", "--no-skip-internal-links", "--reference-links"), - stdin = subprocess.PIPE, stdout = subprocess.PIPE) - page.write(html2markdown.stdin) - html2markdown.stdin.close() - lines = html2markdown.stdout.readlines() - html2markdown.stdout.close() - html2markdown.wait() - - while lines and lines[0].isspace(): - del lines[0] - - fn = fn_base + ".md" - with open(fn, "w") as f: - want_blank = False - for line in lines: - blank = line.isspace() - if want_blank and not blank: - f.write("\n") - if not blank: - f.write(line) - want_blank = blank - sys.stderr.write("Wrote {}\n".format(fn)) - - fn = fn[:-3] + ".wiki" - urllib.urlretrieve(url + "?format=txt", fn) - sys.stderr.write("Wrote {}\n".format(fn)) - - -# XSL transform to extract list of Wiki page URLs from the TOC Wiki page - -xsl_get_toc = lxml.etree.XSLT(lxml.etree.XML('''\ - - - - - - - - - - - - - -''')) - -# XSL transform to extract useful content of a Wiki page. - -# Django generates weird HTML for ordered lists: it sometimes breaks -# up a single ordered list into multiple adjacent
    elements, -# using the @start attribute to try to make the result look like a -# single ordered list. This looks OK in Firefox but confuses the -# bejesus out of both html2markdown and htmldoc. In some cases this is -# probably unavoidable, but most of the uses of this I've seen look -# gratuitous, and are probably the result of code modulararity issues -# in Django. -# -# So we try to clean this up, by merging adjacent
      elements where -# we can. The merge incantation is an adaptation of: -# -# http://stackoverflow.com/questions/1806123/merging-adjacent-nodes-of-same-type-xslt-1-0 -# -# There may be a more efficient way to do this, but I don't think -# we care, and this seems to work. -# -# Original author's explanation: -# -# The rather convoluted XPath expression for selecting the following -# sibling aaa nodes which are merged with the current one: -# -# following-sibling::aaa[ # following 'aaa' siblings -# not(preceding-sibling::*[ # if they are not preceded by -# not(self::aaa) and # a non-'aaa' node -# not(following-sibling::aaa = current()) # after the current node -# ]) -# ] - -xsl_get_page = lxml.etree.XSLT(lxml.etree.XML('''\ - - - - - - - - - - - - - - - - NEW PAGE - - -
      - -
      - - -       - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - _ - - - / - . - - - - - - - - ​ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -       -''')) - -# All the files we want to parse are HTML, so make HTML the default -# parser. In theory the HTML produced by Trac is XHTML thus should -# parse correctly (in fact, better) as XML, but in practice this seems -# not to work properly at the moment, while parsing as HTML does. -# Haven't bothered to figure out why, life is too short. -# -# If you're reading this comment because this script stopped working -# after a Trac upgrade, try commenting out this line to see whether -# things have changed and Trac's HTML now parses better as XML. - -lxml.etree.set_default_parser(lxml.etree.HTMLParser()) - -# Run the main program. -main() diff --git a/references/schema.sql b/references/schema.sql deleted file mode 100644 index 1515dbb..0000000 --- a/references/schema.sql +++ /dev/null @@ -1,177 +0,0 @@ -CREATE TABLE system ( - name text PRIMARY KEY, - value text -); -CREATE TABLE permission ( - username text, - action text, - UNIQUE (username,action) -); -CREATE TABLE auth_cookie ( - cookie text, - name text, - ipnr text, - time integer, - UNIQUE (cookie,ipnr,name) -); -CREATE TABLE session ( - sid text, - authenticated integer, - last_visit integer, - UNIQUE (sid,authenticated) -); -CREATE INDEX session_last_visit_idx ON session (last_visit); -CREATE INDEX session_authenticated_idx ON session (authenticated); -CREATE TABLE session_attribute ( - sid text, - authenticated integer, - name text, - value text, - UNIQUE (sid,authenticated,name) -); -CREATE TABLE cache ( - id integer PRIMARY KEY, - generation integer, - key text -); -CREATE TABLE attachment ( - type text, - id text, - filename text, - size integer, - time integer, - description text, - author text, - ipnr text, - UNIQUE (type,id,filename) -); -CREATE TABLE wiki ( - name text, - version integer, - time integer, - author text, - ipnr text, - text text, - comment text, - readonly integer, - UNIQUE (name,version) -); -CREATE INDEX wiki_time_idx ON wiki (time); -CREATE TABLE repository ( - id integer, - name text, - value text, - UNIQUE (id,name) -); -CREATE TABLE revision ( - repos integer, - rev text, - time integer, - author text, - message text, - UNIQUE (repos,rev) -); -CREATE INDEX revision_repos_time_idx ON revision (repos,time); -CREATE TABLE ticket ( - id integer PRIMARY KEY, - type text, - time integer, - changetime integer, - component text, - severity text, - priority text, - owner text, - reporter text, - cc text, - version text, - milestone text, - status text, - resolution text, - summary text, - description text, - keywords text -); -CREATE INDEX ticket_time_idx ON ticket (time); -CREATE INDEX ticket_status_idx ON ticket (status); -CREATE TABLE ticket_change ( - ticket integer, - time integer, - author text, - field text, - oldvalue text, - newvalue text, - UNIQUE (ticket,time,field) -); -CREATE INDEX ticket_change_ticket_idx ON ticket_change (ticket); -CREATE INDEX ticket_change_time_idx ON ticket_change (time); -CREATE TABLE ticket_custom ( - ticket integer, - name text, - value text, - UNIQUE (ticket,name) -); -CREATE TABLE enum ( - type text, - name text, - value text, - UNIQUE (type,name) -); -CREATE TABLE component ( - name text PRIMARY KEY, - owner text, - description text -); -CREATE TABLE milestone ( - name text PRIMARY KEY, - due integer, - completed integer, - description text -); -CREATE TABLE version ( - name text PRIMARY KEY, - time integer, - description text -); -CREATE TABLE report ( - id integer PRIMARY KEY, - author text, - title text, - query text, - description text -); -CREATE TABLE notify_subscription ( - id integer PRIMARY KEY, - time integer, - changetime integer, - class text, - sid text, - authenticated integer, - distributor text, - format text, - priority integer, - adverb text -); -CREATE INDEX notify_subscription_sid_authenticated_idx ON notify_subscription (sid,authenticated); -CREATE INDEX notify_subscription_class_idx ON notify_subscription (class); -CREATE TABLE notify_watch ( - id integer PRIMARY KEY, - sid text, - authenticated integer, - class text, - realm text, - target text -); -CREATE INDEX notify_watch_sid_authenticated_class_idx ON notify_watch (sid,authenticated,class); -CREATE INDEX notify_watch_class_realm_target_idx ON notify_watch (class,realm,target); -CREATE TABLE node_change ( - id integer PRIMARY KEY, - repos integer, - rev text, - path text, - node_type text, - change_type text, - base_path text, - base_rev text -); -CREATE INDEX node_change_repos_rev_path_idx ON node_change (repos,rev,path); -CREATE INDEX node_change_repos_path_rev_idx ON node_change (repos,path,rev); diff --git a/references/trac-wiki-to-markdown.rb b/references/trac-wiki-to-markdown.rb deleted file mode 100644 index f7d41ae..0000000 --- a/references/trac-wiki-to-markdown.rb +++ /dev/null @@ -1,51 +0,0 @@ -# Untested code snippet from https://gist.github.com/somebox/619537 - -class String - def trac_to_markdown! - gsub!(/\{\{\{([^\n]+?)\}\}\}/, '`\1`') - gsub!(/\{\{\{(.+?)\}\}\}/m){|m| m.each_line.map{|x| "\t#{x}".gsub(/[\{\}]{3}/,'')}.join} - gsub!(/\=\=\=\=\s(.+?)\s\=\=\=\=/, '### \1') - gsub!(/\=\=\=\s(.+?)\s\=\=\=/, '## \1') - gsub!(/\=\=\s(.+?)\s\=\=/, '# \1') - gsub!(/\=\s(.+?)\s\=[\s\n]*/, '') - gsub!(/\[(http[^\s\[\]]+)\s([^\[\]]+)\]/, '[\2](\1)') - gsub!(/\!(([A-Z][a-z0-9]+){2,})/, '\1') - gsub!(/'''(.+)'''/, '*\1*') - gsub!(/''(.+)''/, '_\1_') - gsub!(/^\s\*/, '*') - gsub!(/^\s\d\./, '1.') - - gsub!(/\{\{\{([^\n]+?)\}\}\}/, '`\1`') - gsub!(/'''(.+?)'''/, '**\1**') - gsub!(/''(.+?)''/, '*\1*') - gsub!(/((^\|\|[^\n\r]+\|\|[ \t]*\r?(\n|$))+)/m) do |m| - m = m.each_line.map do |x| - x.gsub(/\t/, ' ') - .gsub(/(\|\|){2,}/){|k| k.gsub(/\|\|/, '|| ')} - .gsub(/ {3,}/, ' ') - end.join - lines = m.each_line.to_a - line1 = lines.shift - line2 = line1.dup.gsub(/[^\n\r\|]/, '-') - lines.unshift(line1, line2) - c = lines.join - c = c.each_line.map do |x| - x.gsub(/\=\s?(.+?)\s?=/, ' \1 ') - .gsub(/\|\|/, '|') - end.join - end - gsub!(/^\{\{\{(.+?)^\}\}\}/m, '```\1```') - gsub!(/\=\=\=\=\s(.+?)\s\=\=\=\=/, '### \1') - gsub!(/\=\=\=\s(.+?)\s\=\=\=/, '## \1') - gsub!(/\=\=\s(.+?)\s\=\=/, '# \1') - gsub!(/\=\s(.+?)\s\=[\s\n]*/, '') - gsub!(/\[(http[^\s\[\]]+)\s([^\[\]]+)\]/, '[\2](\1)') - gsub!(/\!(([A-Z][a-z0-9]+){2,})/, '\1') - gsub!(/^\s\*/, '*') - gsub!(/^\s\d\./, '1.') - end -end - -some_trac = 'my document' - -puts some_trac.trac_to_markdown! diff --git a/references/trac2down.py b/references/trac2down.py deleted file mode 100644 index c66a201..0000000 --- a/references/trac2down.py +++ /dev/null @@ -1,61 +0,0 @@ -#!/usr/bin/env python2 - -# Untested code from https://gist.githubusercontent.com/sgk/1286682/raw/b744dd2e47a68d60373ad39df87cfe8256f517af/trac2down.py - -# vim:set fileencoding=utf-8 sw=2 ai: - -import sqlite3 -import datetime -import re - -SQL = ''' - select - name, version, time, author, text - from - wiki w - where - version = (select max(version) from wiki where name = w.name) -''' - -conn = sqlite3.connect('../trac.db') -result = conn.execute(SQL) -for row in result: - name = row[0] - version = row[1] - time = row[2] - author = row[3] - text = row[4] - - text = re.sub('\r\n', '\n', text) - text = re.sub(r'{{{(.*?)}}}', r'`\1`', text) - def indent4(m): - return '\n ' + m.group(1).replace('\n', '\n ') - text = re.sub(r'(?sm){{{\n(.*?)\n}}}', indent4, text) - text = re.sub(r'(?m)^====\s+(.*?)\s+====$', r'#### \1', text) - text = re.sub(r'(?m)^===\s+(.*?)\s+===$', r'### \1', text) - text = re.sub(r'(?m)^==\s+(.*?)\s+==$', r'## \1', text) - text = re.sub(r'(?m)^=\s+(.*?)\s+=$', r'# \1', text) - text = re.sub(r'^ * ', r'****', text) - text = re.sub(r'^ * ', r'***', text) - text = re.sub(r'^ * ', r'**', text) - text = re.sub(r'^ * ', r'*', text) - text = re.sub(r'^ \d+. ', r'1.', text) - - a = [] - for line in text.split('\n'): - if not line.startswith(' '): - line = re.sub(r'\[(https?://[^\s\[\]]+)\s([^\[\]]+)\]', r'[\2](\1)', line) - line = re.sub(r'\[(wiki:[^\s\[\]]+)\s([^\[\]]+)\]', r'[\2](/\1/)', line) - line = re.sub(r'\!(([A-Z][a-z0-9]+){2,})', r'\1', line) - line = re.sub(r'\'\'\'(.*?)\'\'\'', r'*\1*', line) - line = re.sub(r'\'\'(.*?)\'\'', r'_\1_', line) - a.append(line) - text = '\n'.join(a) - - fp = file('%s.md' % name, 'w') - print >>fp, '' % name - print >>fp, '' % version - print >>fp, '' % datetime.datetime.fromtimestamp(time).strftime('%Y/%m/%d %H:%M:%S') - print >>fp, '' % author - fp.write(text.encode('utf-8')) - fp.close() diff --git a/tools b/tools new file mode 160000 index 0000000..9830a56 --- /dev/null +++ b/tools @@ -0,0 +1 @@ +Subproject commit 9830a562f93005599e09964bcea79c901fb94bb4 diff --git a/trac2md.py b/trac2md.py deleted file mode 100755 index 98374ce..0000000 --- a/trac2md.py +++ /dev/null @@ -1,193 +0,0 @@ -# This started out as https://www.snip2code.com/Snippet/1704331/Convert-trac-markup-to-Markdown/ -# which in turn said "This code mostly taken from patches to pagure_importer by mreynolds". -# Has mutated considerably since then. - -import re -from urllib.parse import quote - -class Trac2Markdown: - - content_linebreak_pattern = re.compile(r"\[\[br\]\]|\\\\", re.I) - camelcase_pattern = re.compile(r"(?:^|(?<=\s))([A-Z][a-z]+[A-Z][a-z][A-Za-z]*)(?:$|(?=\s))") - - wikiheading_patterns = tuple( - (level, re.compile("^{} (.*)[ \t]*=*$".format("=" * level))) - for level in range(1, 7) - ) - - def convert_headers(self, line): - for level_count, header in self.wikiheading_patterns: - try: - level = header.search(line).group(1) - if level: - line = "{} {}".format('#' * level_count, level.rstrip("= \r\t")) - break # No need to check other heading levels - except: - pass # Try the next heading level - return line - - def convert_to_creole(self, m): - # Convert Trac's native link form to Creole's, so that rest of the code only has to deal with one format. - # Creole's is easier to parse and harder to confuse with partially converted Markdown. - - text = m.group(1).strip() - if " " in text: - return "[[{0[0]}|{0[1]}]]".format(text.split(" ", 1)) - elif ":" in text or self.camelcase_pattern.match(text): - return "[[{}]]".format(text) - else: - return m.group(0) - - # Probably most of the non-wiki scheme tests should become a table in an - # extended JSON config file which maps - # - # { "source:fee/fie/foe/fum": "https://git.cryptech.is/blarg/blee/blue" } - - def convert_wikilinks(self, m): - scheme, link, text = [p.strip() if p else p for p in m.groups()] - if text is None: - text = link - if any(link.startswith(q) and link.endswith(q) for q in ('"', "'")): - link = link[1:-1] - if any(text.startswith(q) and text.endswith(q) for q in ('"', "'")): - text = text[1:-1] - if text == link and link.startswith("http") and "://" in link: - return "<{}>".format(link) - elif scheme == "attachment:": - return "[{}]({{attach}}{}/{})".format(text, self.slug, link) - elif scheme in ("source:", "browser:"): - return "[{}]({}/{})".format(text, self.source_url.rstrip("/"), link.lstrip("/")) - elif scheme == "wiki:" or (scheme is None and self.camelcase_pattern.match(link)): - return "[{}]({{filename}}{}.md)".format(text, link) - else: - return "[{}]({})".format(text, link) - - def convert_image(self, m): - text = m.group(1).split(",")[0].strip() - if "://" in text: - return "".format(text) - else: - return "![{}]({{attach}}{}/{})".format(text, self.slug, quote(text, "")) - - def __init__(self, source_url = "https://git.cryptech.is/"): - self.source_url = source_url - self.pattern_actions = ( - - # Convert TracLinks to WikiCreole syntax to simplify remaining processing - (re.compile(r"(?\1"), - - # Convert line breaks -- Markdown spec says linebreak is , who am I to argue? - (re.compile(r"\\\\$"), r" "), - - # Convert bold and italic text (do this last) - (re.compile(r"'''"), r"**"), - (re.compile(r"''"), r"*"), - ) - - def __call__(self, content, slug): - self.slug = slug - - old_content = self.content_linebreak_pattern.sub("\\\\\\\\\n", content).splitlines() - new_content = [] - - code_block = False - in_list = False - in_table = False - nested_level = 0 - prev_indent = 0 - - while old_content: - line = old_content.pop(0).rstrip() - tail = ["\n"] - while "{{{" in line or "}}}" in line: - if "{{{" in line: - code_block = True - line = line.replace("{{{", "```") - if "}}}" in line: - code_block = False - line = line.replace("}}}", "```") - if not code_block: - - # Convert tables. References: - # https://github.github.com/gfm/#tables-extension- - # https://permatrac.noc.ietf.org/wiki/WikiFormatting#Tables - # Table start: line containing "||"; table end: blank line? - # - # Figuring out whether there's a real header line is fun, trac doesn't require one, markdown does. Guess we can - # add a dummy header if no better idea. Markdown requires delimiter line, which we add immediately after the - # header, both appear to be mandatory. Trac can have label cells anywhere, not just in header, might need to - # add "*" to those or just ignore the issue. Justification we can sort of figure out from the header, - # if the rows do anything different, ouch, because markdown specifies in delimiter line. - # - # Might do something clever with the "=" markers and alignment, start with just getting the basic table - # structure to something markdown will believe. - - if line.strip().startswith("||"): - line = line.replace("=|", "|").replace("|=", "|") - line = line.replace("||", "|") - if not in_table: - tail.append("|---" * (line.count("|") - 1) + "|\n") - in_table = True - elif in_table and not line.strip().startswith("||"): - new_content.append("\n") - in_table = False - - # - # Convert bullet lists. The start and end of a list needs an empty line. - # - nested_line = line.lstrip(' ') - if nested_line.startswith('- ') or nested_line.startswith('* '): - if not in_list: - new_content.append("\n") - nested_level = 0 - prev_indent = 0 - in_list = True - indent = len(line) - len(nested_line) - text_indent = len(line) - len(nested_line[1:].lstrip()) - if indent > prev_indent: - nested_level += 1 - elif indent < prev_indent: - nested_level -= 1 - prev_indent = indent - line = ' ' * nested_level + nested_line - elif in_list and len(line) < len(nested_line) + text_indent: - new_content.append("\n") - in_list = False - nested_level = 0 - prev_indent = 0 - text_indent = 0 - - # Convert headers - line = self.convert_headers(line) - - # Rest is regexp-driven conversions - for pattern, action in self.pattern_actions: - line = pattern.sub(action, line) - - new_content.append(line) - new_content.extend(tail) - - del self.slug - - return "".join(new_content) -- cgit v1.2.3