def saveEntry(self, thisEntry: BaseEntry, thisHash: str, prevHash: str, nextHash: str) -> None:
dpath = join(self._filename, thisHash[:2])
makeDir(dpath)
with open(
join(dpath, thisHash[2:]),
"w",
encoding=self._encoding,
) as toFile:
nextPath = self.hashToPath(nextHash) if nextHash else "END"
if self._havePrevLink:
prevPath = self.hashToPath(prevHash) if prevHash else "START"
header = prevPath + " " + nextPath
else:
header = nextPath
toFile.write("\n".join([
header,
escapeNTB(thisEntry.s_word, bar=False),
thisEntry.defi,
]))
def write(self) -> "Generator[None, BaseEntry, None]":
encoding = self._encoding
resources = self._resources
max_file_size = self._max_file_size
filename_format = self._filename_format
escape_defi = self._escape_defi
wordSep = ' <font color="red">|</font> '
initFileSizeMax = 100
glos = self._glos
filename = self._filename
self._encoding = encoding
self._filename_format = filename_format
entry_url_fmt = glos.getInfo("entry_url")
def getEntryWebLink(entry) -> str:
if not entry_url_fmt:
return ""
url = entry_url_fmt.format(word=html.escape(entry.l_word[0]))
return f'{nbsp}<a class="no_ul" href="{url}">🌏</a>'
# from math import log2, ceil
# maxPosHexLen = int(ceil(log2(max_file_size) / 4))
indexTxtFileObj = open(
join(filename, "index.txt"),
mode="w",
encoding="utf-8",
)
linksTxtFileObj = open(
join(filename, "links.txt"),
mode="w",
encoding="utf-8",
)
title = glos.getInfo("name")
style = ""
if self._dark:
style = darkStyle
if self._css:
cssLink = '<link rel="stylesheet" href="style.css" />'
else:
cssLink = ""
header = (
'<!DOCTYPE html>\n'
'<html><head>'
f'<title>{{pageTitle}}</title>'
f'<meta charset="{encoding}">'
f'<style type="text/css">{style}{{customStyle}}</style>{cssLink}'
'</meta></head><body>\n')
def pageHeader(n: int):
return header.format(
pageTitle=f"Page {n} of {title}",
customStyle="",
)
def navBar() -> str:
links = []
if len(self._filenameList) > 1:
links.append(
f'<a href="./{self._filenameList[-2]}">◀</a>')
links.append(f'<a href="./{self.getNextFilename()}">▶</a>')
links.append(f'<a href="./info.html">ℹ️</a></div>')
return ('<div style="text-align: center; font-size: 2.5em;">' +
f'{nbsp}{nbsp}{nbsp}'.join(links) + '</div>')
tailSize = len(self._tail.encode(encoding))
if max_file_size < len(header) + tailSize:
raise ValueError(f"max_file_size={max_file_size} is too small")
max_file_size -= tailSize
if not isdir(self._filename):
os.mkdir(self._filename)
fileObj = self.nextFile()
fileObj.write(pageHeader(0))
fileObj.write(navBar())
re_fixed_link = re.compile(
r'<a (?:[^<>]*? )?href="#([^<>"]+?)">[^<>]+?</a>',
re.I,
)
linkTargetSet = set()
def replaceBword(text) -> str:
return text.replace(
' href="bword://',
' href="#',
)
def addLinks(text: str, pos: int) -> str:
for m in re_fixed_link.finditer(text):
if ' class="entry_link"' in m.group(0):
continue
if m.group(0).count("href=") != 1:
log.error(f"unexpected match: {m.group(0)}")
target = html.unescape(m.group(1))
linkTargetSet.add(target)
start = m.start()
b_start = len(text[:start].encode(encoding))
b_size = len(text[start:m.end()].encode(encoding))
linksTxtFileObj.write(f"{escapeNTB(target)}\t"
f"{len(self._filenameList)-1}\t"
f"{hex(pos+b_start)[2:]}\t"
f"{hex(b_size)[2:]}\n")
linksTxtFileObj.flush()
self.writeInfo(filename, header)
_word_title = self._word_title
resDir = self._resDir
entryIndex = -1
while True:
entryIndex += 1
entry = yield
if entry is None:
break
if entry.isData():
if resources:
entry.save(resDir)
continue
if entry.defi.startswith('<!DOCTYPE html>') and defiFormat != "h":
log.error(f"bad defiFormat={defiFormat}")
defiFormat = "h"
entry.detectDefiFormat()
entry.stripFullHtml()
defi = entry.defi
defiFormat = entry.defiFormat
if defiFormat == "m":
defi = html.escape(defi)
if "\n" in defi:
# could be markdown or unformatted plaintext
# FIXME: this changes the font to a monospace
defi = f'<pre>{defi}</pre>'
elif defiFormat == "h":
if escape_defi:
defi = html.escape(defi)
defi = defi.replace(' src="./', ' src="./res/')
entryId = f"entry{entryIndex}"
if _word_title:
words = [html.escape(word) for word in entry.l_word]
title = glos.wordTitleStr(
wordSep.join(words),
sample=entry.l_word[0],
_class="headword",
)
if not title:
title = f'Entry {entryIndex}'
# entry_link_sym = "¶"
entry_link_sym = "🔗"
text = (f'<div id="{entryId}">{title}{nbsp}{nbsp}'
f'<a class="no_ul" class="entry_link" href="#{entryId}">'
f'{entry_link_sym}</a>'
f'{getEntryWebLink(entry)}'
f"<br>\n{defi}"
'</div>\n'
'<hr>\n')
pos = fileObj.tell()
if pos > initFileSizeMax:
if pos > max_file_size - len(text.encode(encoding)):
fileObj = self.nextFile()
fileObj.write(pageHeader(len(self._filenameList) - 1))
fileObj.write(navBar())
pos = fileObj.tell()
tmpFilename = escapeNTB(self._filenameList[-1])
for word in entry.l_word:
indexTxtFileObj.write(f"{entryIndex}\t"
f"{escapeNTB(word)}\t"
f"{tmpFilename}\t"
f"{pos}\n")
del tmpFilename
text = replaceBword(text)
addLinks(text, pos)
fileObj.write(text)
fileObj.close()
self._fileObj = None
indexTxtFileObj.close()
if linkTargetSet:
log.info(f"{len(linkTargetSet)} link targets found")
log.info("Fixing links, please wait...")
self.fixLinks(linkTargetSet)
os.remove(join(filename, "links.txt"))
def write(self) -> Generator[None, "BaseEntry", None]:
encoding = self._encoding
resources = self._resources
max_file_size = self._max_file_size
filename_format = self._filename_format
escape_defi = self._escape_defi
wordSep = ' <font color="red">|</font> '
initFileSizeMax = 100
glos = self._glos
filename = self._filename
self._encoding = encoding
self._filename_format = filename_format
entry_url_fmt = glos.getInfo("entry_url")
def getEntryWebLink(entry) -> str:
if not entry_url_fmt:
return ""
url = entry_url_fmt.format(word=html.escape(entry.l_word[0]))
return f' <a class="no_ul" href="{url}">🌏</a>'
# from math import log2, ceil
# maxPosHexLen = int(ceil(log2(max_file_size) / 4))
indexTxtFileObj = open(
join(filename, "index.txt"),
mode="w",
encoding="utf-8",
)
linksTxtFileObj = open(
join(filename, "links.txt"),
mode="w",
encoding="utf-8",
)
title = glos.getInfo("name")
style = ""
if self._dark:
style = darkStyle
header = ('<!DOCTYPE html>\n'
'<html><head>'
f'<title>{{pageTitle}}</title>'
f'<meta charset="{encoding}">'
f'<style type="text/css">{style}{{customStyle}}</style>'
'</meta></head><body>\n')
def pageHeader(n: int):
return header.format(
pageTitle=f"Page {n} of {title}",
customStyle="",
)
def stripEntryFullHtml(entry, defi: str) -> str:
word = entry.s_word
i = defi.find('<body')
if i == -1:
log.error(f"<body not found: word={word}")
return defi
defi = defi[i + 5:]
i = defi.find('>')
if i == -1:
log.error(f"'>' after <body not found: word={word}")
return defi
defi = defi[i + 1:]
i = defi.find('</body')
if i == -1:
log.error(f"</body close not found: word={word}")
return defi
defi = defi[:i]
return defi
def navBar() -> str:
links = []
if len(self._filenameList) > 1:
links.append(
f'<a href="./{self._filenameList[-2]}">◀</a>')
links.append(f'<a href="./{self.getNextFilename()}">▶</a>')
links.append(f'<a href="./info.html">ℹ️</a></div>')
return ('<div style="text-align: center; font-size: 2.5em;">' +
' '.join(links) + '</div>')
tailSize = len(self._tail.encode(encoding))
if max_file_size < len(header) + tailSize:
raise ValueError(f"max_file_size={max_file_size} is too small")
max_file_size -= tailSize
if not isdir(self._filename):
os.mkdir(self._filename)
fileObj = self.nextFile()
fileObj.write(pageHeader(0))
fileObj.write(navBar())
re_fixed_link = re.compile(
r'<a (?:[^<>]*? )?href="#([^<>"]+?)">[^<>]+?</a>',
re.I,
)
linkTargetSet = set()
def replaceBword(text) -> str:
return text.replace(
' href="bword://',
' href="#',
)
def addLinks(text: str, pos: int) -> str:
for m in re_fixed_link.finditer(text):
if ' class="entry_link"' in m.group(0):
continue
if m.group(0).count("href=") != 1:
log.error(f"unexpected match: {m.group(0)}")
target = html.unescape(m.group(1))
linkTargetSet.add(target)
start = m.start()
b_start = len(text[:start].encode(encoding))
b_size = len(text[start:m.end()].encode(encoding))
linksTxtFileObj.write(f"{escapeNTB(target)}\t"
f"{len(self._filenameList)-1}\t"
f"{hex(pos+b_start)[2:]}\t"
f"{hex(b_size)[2:]}\n")
linksTxtFileObj.flush()
self.writeInfo(filename, header)
defiHasHeadwords = glos.getInfo("definition_has_headwords") == "True"
resDir = self._resDir
entryIndex = -1
while True:
entryIndex += 1
entry = yield
if entry is None:
break
if entry.isData():
if resources:
entry.save(resDir)
continue
defi = entry.defi
defiFormat = entry.defiFormat
if defi.startswith('<!DOCTYPE html>'):
if defiFormat != "h":
log.error(f"bad defiFormat={defiFormat}")
defiFormat = "h"
defi = stripEntryFullHtml(entry, defi)
if defiFormat == "m":
defi = defi.replace("\n", "<br>")
if escape_defi:
defi = html.escape(defi)
entryId = f"entry{entryIndex}"
if defiHasHeadwords:
headwords = f'Entry {entryIndex}'
else:
headwords = f'<b class="headword">{wordSep.join(entry.l_word)}</b>'
text = (
f'<div id="{entryId}">{headwords} '
f'<a class="no_ul" class="entry_link" href="#{entryId}">🔗</a>'
f'{getEntryWebLink(entry)}'
f"<br>\n{defi}"
'</div>\n'
'<hr>\n')
pos = fileObj.tell()
if pos > initFileSizeMax:
if pos > max_file_size - len(text.encode(encoding)):
fileObj = self.nextFile()
fileObj.write(pageHeader(len(self._filenameList) - 1))
fileObj.write(navBar())
pos = fileObj.tell()
tmpFilename = escapeNTB(self._filenameList[-1])
for word in entry.l_word:
indexTxtFileObj.write(f"{entryIndex}\t"
f"{escapeNTB(word)}\t"
f"{tmpFilename}\t"
f"{pos}\n")
del tmpFilename
text = replaceBword(text)
addLinks(text, pos)
fileObj.write(text)
fileObj.close()
self._fileObj = None
indexTxtFileObj.close()
if linkTargetSet:
log.info(f"\n{len(linkTargetSet)} link targets found")
log.info("Fixing links, please wait...")
self.fixLinks(linkTargetSet)
os.remove(join(filename, "links.txt"))
