evennia/docs/pylib/copy_from_wiki.py
2022-11-10 22:21:12 +01:00

280 lines
8.5 KiB
Python

#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
Copy data from old Evennia github Wiki to static files.
Prepare files for mkdoc. This assumes evennia.wiki is cloned
to a folder at the same level as the evennia repo.
Just run this to update everything.
We also need to build the toc-tree and should do so automatically for now.
"""
import datetime
import glob
import re
_RE_MD_LINK = re.compile(r"\[(?P<txt>[\w -\[\]]+?)\]\((?P<url>.+?)\)", re.I + re.S + re.U)
_RE_REF_LINK = re.compile(r"\[[\w -\[\]]*?\]\(.+?\)", re.I + re.S + re.U)
_RE_CLEAN = re.compile(r"\|-+?|-+\|", re.I + re.S + re.U)
_IGNORE_FILES = (
"_Sidebar.md",
# "Wiki-Index.md"
)
_INDEX_PREFIX = f"""
# VERSION WARNING
> This is the experimental static v0.9 documentation of Evennia, _automatically_ generated from the
> [evennia wiki](https://github.com/evennia/evennia/wiki/) at {datetime.datetime.now()}.
> There are known conversion issues which will _not_ be addressed in this version - refer to
> the original wiki if you have trouble.
>
> Manual conversion and cleanup will instead happen during development of the upcoming v1.0
> version of this static documentation.
"""
_WIKI_DIR = "../../../evennia.wiki/"
_INFILES = [
path
for path in sorted(glob.glob(_WIKI_DIR + "/*.md"))
if path.rsplit("/", 1)[-1] not in _IGNORE_FILES
]
_FILENAMES = [path.rsplit("/", 1)[-1] for path in _INFILES]
_FILENAMES = [path.split(".", 1)[0] for path in _FILENAMES]
_FILENAMESLOW = [path.lower() for path in _FILENAMES]
_OUTDIR = "../source/"
_OLD_WIKI_URL = "https://github.com/evennia/evennia/wiki/"
_OLD_WIKI_URL_LEN = len(_OLD_WIKI_URL)
_CODE_PREFIX = "github:"
_API_PREFIX = "api:"
_CUSTOM_LINK_REMAP = {
"CmdSets": "Command-Sets",
"CmdSet": "Command-Sets",
"Cmdsets": "Command-Sets",
"CommandSet": "Command-Sets",
"batch-code-processor": "Batch-Code-Processor",
"Batch-code-processor": "Batch-Code-Processor",
"batch-command-processor": "Batch-Command-Processor",
"Batch-command-processor": "Batch-Command-Processor",
"evennia-API": "Evennia-API",
"Channels": "Communications#Channels",
"Comms": "Communications",
"typeclass": "Typeclasses",
"Home": "index",
"Help-system": "Help-System",
"Using-Mux-as-a-Standard": "Using-MUX-as-a-Standard",
"Building-quickstart": "Building-Quickstart",
"Adding-Object-Typeclass-tutorial": "Adding-Object-Typeclass-Tutorial",
"EvTable": _API_PREFIX + "evennia.utils#module-evennia.utils.evtable",
}
# complete reference remaps
_REF_REMAP = {
"[![Getting Started][icon_new]](Getting-Started)": "![Getting Started][icon_new]",
"[![Admin Docs][icon_admin]](Administrative-Docs)": "![Admin Docs][icon_admin]",
"[![Builder Docs][icon_builder]](Builder-Docs)": "![Builder Docs][icon_builder]",
"[![Developer-Central][icon_devel]](Developer-Central)": "![Developer-Central][icon_devel]",
"[![tutorial][icon_tutorial]](Tutorials)": "![Tutorials][icon_tutorial]",
"[![API][icon_api]](evennia)": "![API][icon_api]",
"[](Wiki-front-page.)": "",
}
# absolute links (mainly github links) that should not be converted. This
# should be given without any #anchor.
_ABSOLUTE_LINK_SKIP = (
# "https://github.com/evennia/evennia/wiki/feature-request",
)
# specific references tokens that should be ignored. Should be given
# without any #anchor.
_REF_SKIP = (
"[5](Win)",
"[6](Win)",
"[7](Win)",
"[10](Win)",
"[11](Mac)",
"[13](Win)",
"[14](IOS)",
"[15](IOS)",
"[16](Andr)",
"[17](Andr)",
"[18](Unix)",
"[21](Chrome)",
# these should be checked
"[EvTable](EvTable)",
"[styled](OptionStyles)",
"[Inputfunc](Inputfunc)",
"[online documentation wiki](index)",
"[online documentation](index)",
"[Accounts](Account)",
"[Session](Session)",
"[Inputfuncs](Inputfunc)",
)
_CURRENT_TITLE = ""
def _sub_remap(match):
"""Total remaps"""
ref = match.group(0)
if ref in _REF_REMAP:
new_ref = _REF_REMAP[ref]
print(f" Replacing reference {ref} -> {new_ref}")
return new_ref
return ref
def _sub_link(match):
mdict = match.groupdict()
txt, url_orig = mdict["txt"], mdict["url"]
url = url_orig
# if not txt:
# # the 'comment' is not supported by Mkdocs
# return ""
print(f" [{txt}]({url})")
url = _CUSTOM_LINK_REMAP.get(url, url)
url, *anchor = url.rsplit("#", 1)
if url in _ABSOLUTE_LINK_SKIP:
url += ("#" + anchor[0]) if anchor else ""
return f"[{txt}]({url})"
if url.startswith("evennia"):
print(f" Convert evennia url {url} -> {_CODE_PREFIX + url}")
url = _API_PREFIX + url
if url.startswith(_OLD_WIKI_URL):
# old wiki is an url on the form https://<wikiurl>/wiki/TextTags#header
# we don't refer to the old wiki but use internal mapping.
if len(url) != len(_OLD_WIKI_URL):
url_conv = url[_OLD_WIKI_URL_LEN:]
url_conv = re.sub(r"%20", "-", url_conv)
if url_conv.endswith("/_edit"):
# this is actually a bug in the wiki format
url_conv = url_conv[:-6]
if url_conv.startswith("evennia"):
# this is an api link
url_conv = _CODE_PREFIX + url_conv
print(f" Converting wiki-url: {url} -> {url_conv}")
url = url_conv
if not url and anchor:
# this happens on same-file #labels in wiki
url = _CURRENT_TITLE
if url not in _FILENAMES and not url.startswith("http") and not url.startswith(_CODE_PREFIX):
url_cap = url.capitalize()
url_plur = url[:-3] + "s" + ".md"
url_cap_plur = url_plur.capitalize()
link = f"[{txt}]({url})"
if link in _REF_SKIP:
url = link
elif url_cap in _FILENAMES:
print(f" Replacing (capitalized): {url.capitalize()}")
url = url_cap
elif url_plur in _FILENAMES:
print(f" Replacing (pluralized): {url + 's'}")
url = url_plur
elif url_cap_plur in _FILENAMES:
print(f" Replacing (capitalized, pluralized): {url.capitalize() + 's'}")
url = url_cap_plur
elif url.lower() in _FILENAMESLOW:
ind = _FILENAMESLOW.index(url.lower())
alt = _FILENAMES[ind]
print(f" Replacing {url} with different cap: {alt}")
url = alt
# print(f"\nlink {link} (orig: [{txt}]({url_orig})) found no file match")
# inp = input("Enter alternate url (return to keep old): ")
# if inp.strip():
# url = inp.strip()
if anchor:
url += "#" + anchor[0]
return f"[{txt}]({url})"
def create_toctree(files):
with open("../source/toc.md", "w") as fil:
fil.write("# Toc\n")
for path in files:
filename = path.rsplit("/", 1)[-1]
ref = filename.rsplit(".", 1)[0]
linkname = ref.replace("-", " ")
if ref == "Home":
ref = "index"
fil.write(f"\n* [{linkname}]({ref}.md)")
def convert_links(files, outdir):
global _CURRENT_TITLE
for inpath in files:
is_index = False
outfile = inpath.rsplit("/", 1)[-1]
if outfile == "Home.md":
outfile = "index.md"
is_index = True
outfile = _OUTDIR + outfile
title = inpath.rsplit("/", 1)[-1].split(".", 1)[0].replace("-", " ")
print(f"Converting links in {inpath} -> {outfile} ...")
with open(inpath) as fil:
text = fil.read()
if is_index:
text = _INDEX_PREFIX + text
lines = text.split("\n")
lines = (
lines[:-11]
+ [" - The [TOC](toc) lists all regular documentation pages.\n\n"]
+ lines[-11:]
)
text = "\n".join(lines)
_CURRENT_TITLE = title.replace(" ", "-")
text = _RE_CLEAN.sub("", text)
text = _RE_REF_LINK.sub(_sub_remap, text)
text = _RE_MD_LINK.sub(_sub_link, text)
text = (
text.split("\n")[1:]
if text.split("\n")[0].strip().startswith("[]")
else text.split("\n")
)
text = "\n".join(text)
if not is_index:
text = f"# {title}\n\n{text}"
with open(outfile, "w") as fil:
fil.write(text)
if __name__ == "__main__":
print("This should not be run on develop files, it would overwrite changes.")
# create_toctree(_INFILES)
# convert_links(_INFILES, _OUTDIR)