Skip to content

Commit dea623b

Browse files
q1zhenq1zhen
committed
fix: updated make script
1 parent 2bfb9ea commit dea623b

File tree

1 file changed

+198
-142
lines changed

1 file changed

+198
-142
lines changed

‎scripts/make.py‎

Lines changed: 198 additions & 142 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,5 @@
11
import os
2+
import re
23
import sys
34
import shutil
45
import hashlib
@@ -14,164 +15,219 @@
1415
bch_size = 5
1516

1617
class File():
17-
def __init__(self, path: str):
18-
self.path = path
19-
self.dir = os.path.dirname(self.path)
20-
self.filename = os.path.basename(self.path)
21-
os.makedirs(self.dir, exist_ok=True)
22-
def read(self):
23-
with open(self.path, "r", encoding="utf-8") as f:
24-
content = f.read()
25-
return content
26-
def readlines(self):
27-
with open(self.path, "r", encoding="utf-8") as f:
28-
lines = f.readlines()
29-
return lines
30-
def write(self, content: str):
31-
with open(self.path, "w+", encoding="utf-8") as f:
32-
f.write(content)
33-
def writelines(self, lines: list):
34-
with open(self.path, "w+", encoding="utf-8") as f:
35-
f.writelines(lines)
18+
def __init__(self, path: str):
19+
self.path = path
20+
self.dir = os.path.dirname(self.path)
21+
self.filename = os.path.basename(self.path)
22+
os.makedirs(self.dir, exist_ok=True)
23+
def read(self):
24+
with open(self.path, "r", encoding="utf-8") as f:
25+
content = f.read()
26+
return content
27+
def readlines(self):
28+
with open(self.path, "r", encoding="utf-8") as f:
29+
lines = f.readlines()
30+
return lines
31+
def write(self, content: str):
32+
with open(self.path, "w+", encoding="utf-8") as f:
33+
f.write(content)
34+
def writelines(self, lines: list):
35+
with open(self.path, "w+", encoding="utf-8") as f:
36+
f.writelines(lines)
3637

3738
def hash_str(s):
38-
return hashlib.sha256(s.encode()).hexdigest()
39+
return hashlib.sha256(s.encode()).hexdigest()
3940

4041
def metaext(src):
41-
print(" Extracting metadata:")
42-
lines = File(src).readlines()
43-
global meta
44-
meta = {}
45-
inmd = False
46-
fpos = -1
47-
for line in lines:
48-
fpos += 1
49-
if line.strip() == "---":
50-
lines[fpos] = ""
51-
if inmd: break
52-
else: inmd = True
53-
else:
54-
if ":" not in line: continue
55-
key, val = line.split(":", 1)
56-
lines[fpos] = ""
57-
meta[key.strip()] = val.strip()
58-
if len(meta):
59-
print("title: %s; author: %s; date: %s" % (meta.get("title"), meta.get("author"), meta.get("date")))
60-
File(src).writelines(lines)
61-
else:
62-
print(" Error: no metadata found")
42+
print(" Extracting metadata:")
43+
lines = File(src).readlines()
44+
global meta
45+
meta = {}
46+
inmd = False
47+
fpos = -1
48+
for line in lines:
49+
fpos += 1
50+
if line.strip() == "---":
51+
lines[fpos] = ""
52+
if inmd: break
53+
else: inmd = True
54+
else:
55+
if ":" not in line: continue
56+
key, val = line.split(":", 1)
57+
lines[fpos] = ""
58+
meta[key.strip()] = val.strip()
59+
if len(meta):
60+
print("title: %s; author: %s; date: %s" % (meta.get("title"), meta.get("author"), meta.get("date")))
61+
File(src).writelines(lines)
62+
else:
63+
print(" Error: no metadata found")
64+
65+
def sanitize_meta(value: str) -> str:
66+
"""Sanitize a metadata value for safe LaTeX insertion."""
67+
if value is None:
68+
return ""
69+
70+
sanitized = value.strip()
71+
if sanitized.startswith(('"', "'")) and sanitized.endswith(('"', "'")):
72+
sanitized = sanitized[1:-1]
73+
74+
replacements = {
75+
"\\": r"\\textbackslash{}",
76+
"{": r"\\{",
77+
"}": r"\\}",
78+
"$": r"\\$",
79+
"&": r"\\&",
80+
"#": r"\\#",
81+
"_": r"\\_",
82+
"%": r"\\%",
83+
"^": r"\\textasciicircum{}",
84+
"~": r"\\textasciitilde{}",
85+
}
86+
87+
for needle, replacement in replacements.items():
88+
sanitized = sanitized.replace(needle, replacement)
89+
90+
return sanitized
6391

6492
def metainj(dst):
65-
print(" Injecting metadata:")
66-
manu = File(dst).readlines()
67-
manu.insert(0, "\\title{%s}\n\\author{%s}\n\\date{%s}\n\\maketitle\n" % (meta.get("title"), meta.get("author"), meta.get("date")))
68-
File(dst).writelines(manu)
93+
print(" Injecting metadata:")
94+
manu = File(dst).readlines()
95+
title = sanitize_meta(meta.get("title"))
96+
author = sanitize_meta(meta.get("author"))
97+
date = sanitize_meta(meta.get("date"))
98+
manu.insert(0, f"\\title{{{title}}}\n\\author{{{author}}}\n\\date{{{date}}}\n\\maketitle\n")
99+
File(dst).writelines(manu)
100+
101+
def texpost(dst):
102+
"""Apply post-processing to generated TeX content."""
103+
content = File(dst).read()
104+
def _sanitize_text(match):
105+
inner = match.group(1)
106+
inner = inner.replace("\\", "\\textbackslash{}")
107+
inner = inner.replace("\\textbackslash{}\\textbackslash{}", "\\textbackslash{}")
108+
inner = re.sub(r"(?<!\\)_", r"\\_", inner)
109+
return f"\\text{{{inner}}}"
110+
111+
content = re.sub(r"\\text\{([^}]*)\}", _sanitize_text, content)
112+
content = re.sub(r"\\colon(?=\w)", r"\\colon ", content)
113+
content = re.sub(r"\\nabla(?=\w)", r"\\nabla ", content)
114+
content = re.sub(r"\\oplus(?=\w)", r"\\oplus ", content)
115+
content = re.sub(
116+
r"\\(Delta|delta|Gamma|gamma|Theta|theta|Lambda|lambda|Xi|xi|Pi|pi|Sigma|sigma|Upsilon|upsilon|Phi|phi|Psi|psi|Omega|omega)(?=[A-Za-z])",
117+
r"\\\1 ",
118+
content,
119+
)
120+
content = re.sub(r"\\cdot(?=\w)", r"\\cdot ", content)
121+
content = re.sub(r"\\times(?=\w)", r"\\times ", content)
122+
content = re.sub(r"\\log(?=\w)", r"\\log ", content)
123+
File(dst).write(content)
69124

70125
def texcomp(drv):
71-
print(" Generating PDF:")
72-
shutil.copy(utl_dir + drv, tmp_dir + "index.ltx")
73-
for fonts in os.listdir(fnt_dir):
74-
shutil.copy(fnt_dir + fonts, tmp_dir + fonts)
75-
for macro in os.listdir(sty_dir):
76-
shutil.copy(sty_dir + macro, tmp_dir + macro)
77-
pwd = os.getcwd()
78-
os.chdir(tmp_dir)
79-
os.system("lualatex index.ltx --interaction=batchmode")
80-
os.chdir(pwd)
126+
print(" Generating PDF:")
127+
shutil.copy(utl_dir + drv, tmp_dir + "index.ltx")
128+
for fonts in os.listdir(fnt_dir):
129+
shutil.copy(fnt_dir + fonts, tmp_dir + fonts)
130+
for macro in os.listdir(sty_dir):
131+
shutil.copy(sty_dir + macro, tmp_dir + macro)
132+
pwd = os.getcwd()
133+
os.chdir(tmp_dir)
134+
os.system("lualatex index.ltx --interaction=batchmode")
135+
os.chdir(pwd)
81136

82137
def pdfgenr(post):
83-
shutil.copytree(src_dir + post, tmp_dir, dirs_exist_ok=True)
84-
metaext(tmp_dir + "index.md")
85-
print(" Converting:")
86-
os.system(f"{utl_dir}md2tex {tmp_dir}index.md {tmp_dir}index.tex")
87-
metainj(tmp_dir + "index.tex")
88-
texcomp("drvpst.ltx")
89-
shutil.copy(tmp_dir + "index.tex", pbl_dir + post + "/index.tex")
90-
shutil.copy(tmp_dir + "index.pdf", pbl_dir + post + "/index.pdf")
91-
shutil.rmtree(tmp_dir)
138+
shutil.copytree(src_dir + post, tmp_dir, dirs_exist_ok=True)
139+
metaext(tmp_dir + "index.md")
140+
print(" Converting:")
141+
os.system(f"{utl_dir}md2tex {tmp_dir}index.md {tmp_dir}index.tex")
142+
metainj(tmp_dir + "index.tex")
143+
texpost(tmp_dir + "index.tex")
144+
texcomp("drvpst.ltx")
145+
shutil.copy(tmp_dir + "index.tex", pbl_dir + post + "/index.tex")
146+
shutil.copy(tmp_dir + "index.pdf", pbl_dir + post + "/index.pdf")
147+
shutil.rmtree(tmp_dir)
92148

93149
def post():
94-
# Compiles each post, specifically converts the .md file to a .tex file
95-
# and compiles the .tex, finally moving .tex and .pdf to the public dir
96-
cwd = os.getcwd()
97-
print(f" Making directory: {cwd}")
98-
os.chdir(utl_dir)
99-
os.system("make")
100-
os.chdir(cwd)
101-
# For each of the posts in the source directory
102-
for post in sorted(os.listdir(src_dir)):
103-
# Get hash of the markdown file content
104-
hsh = hash_str(File(src_dir + post + "/index.md").read())
105-
# If it is already compiled and the hash matches
106-
if all([
107-
os.path.exists(pbl_dir + post + p) for p in ["/index.tex", "/index.pdf", "/sha256"]
108-
]) and File(pbl_dir + post + "/sha256").read() == hsh:
109-
print(f" Skipping post: {post} #{hsh}")
110-
continue
111-
print(f" Processing post: {post}")
112-
# Write markdown hash to file `sha256`
113-
File(pbl_dir + post + "/sha256").write(hsh)
114-
# Compile PDF
115-
pdfgenr(post)
116-
# Cleaning
117-
os.chdir(utl_dir)
118-
print(" Cleaning up:")
119-
os.system("make clean")
150+
# Compiles each post, specifically converts the .md file to a .tex file
151+
# and compiles the .tex, finally moving .tex and .pdf to the public dir
152+
cwd = os.getcwd()
153+
print(f" Making directory: {cwd}")
154+
os.chdir(utl_dir)
155+
os.system("make")
156+
os.chdir(cwd)
157+
# For each of the posts in the source directory
158+
for post in sorted(os.listdir(src_dir)):
159+
# Get hash of the markdown file content
160+
hsh = hash_str(File(src_dir + post + "/index.md").read())
161+
# If it is already compiled and the hash matches
162+
if all([
163+
os.path.exists(pbl_dir + post + p) for p in ["/index.tex", "/index.pdf", "/sha256"]
164+
]) and File(pbl_dir + post + "/sha256").read() == hsh:
165+
print(f" Skipping post: {post} #{hsh}")
166+
continue
167+
print(f" Processing post: {post}")
168+
# Write markdown hash to file `sha256`
169+
File(pbl_dir + post + "/sha256").write(hsh)
170+
# Compile PDF
171+
pdfgenr(post)
172+
# Cleaning
173+
os.chdir(utl_dir)
174+
print(" Cleaning up:")
175+
os.system("make clean")
120176

121177
def batch():
122-
# This function compiles several files sequentially into one batch version
123-
# The number of files in each batch are defined in `bch_size`, remaining
124-
# files not reaching that number will not be included as a new batch.
125-
cwd = os.getcwd()
126-
print(f" Making directory: {cwd}")
127-
# Reading all posts (date strings)
128-
posts = sorted(os.listdir(src_dir))
129-
# Extracting batch ID and hash from preexisting batch directory
130-
compiled = [i.split(".")[0].split("_") for i in sorted(os.listdir(bch_dir))]
131-
compiled_hsh = {
132-
int(i[1]): i[2] for i in compiled
133-
}
134-
# Generating each batch
135-
for bch_id, bch_start in enumerate(range(0, len(posts), bch_size)):
136-
# For remaining files at the end not reaching the size of a batch
137-
if bch_start + bch_size > len(posts): break
138-
# Current range of indices and hash of date strings
139-
bch_range = list(range(bch_start, bch_start + bch_size))
140-
# Join contents of each post to generate hash
141-
hsh = hash_str("".join([
142-
File(f"{src_dir}{posts[i]}/index.md").read() for i in bch_range
143-
]))[-6:]
144-
existing_hsh = compiled_hsh.get(bch_id)
145-
# If this batch is already present
146-
if existing_hsh:
147-
if existing_hsh == hsh:
148-
print(f" Skipping batch: {bch_id} #{hsh}")
149-
continue
150-
else:
151-
print(f" Removing obsolete: {bch_id} #{existing_hsh} -> #{hsh}")
152-
os.remove(f"{bch_dir}compilation_{bch_id}_{existing_hsh}.pdf")
153-
print(f" Processing batch: {bch_id} #{hsh}")
154-
filename = f"compilation_{bch_id}_{hsh}"
155-
# Writing index.tex to be compiled
156-
File(f"{tmp_dir}index.tex").writelines([
157-
# Title
158-
"\\mlytitle{" + f"c13n \\#{bch_id}" + "}",
159-
*[
160-
# Dump file contents from each post directory
161-
File(f"{pbl_dir}{posts[i]}/index.tex").read()
162-
for i in bch_range
163-
]
164-
])
165-
# Compiling and cleaning
166-
texcomp("drvmly.ltx")
167-
shutil.copy(tmp_dir + "index.pdf", bch_dir + filename.lower() + ".pdf")
168-
shutil.rmtree(tmp_dir)
178+
# This function compiles several files sequentially into one batch version
179+
# The number of files in each batch are defined in `bch_size`, remaining
180+
# files not reaching that number will not be included as a new batch.
181+
cwd = os.getcwd()
182+
print(f" Making directory: {cwd}")
183+
# Reading all posts (date strings)
184+
posts = sorted(os.listdir(src_dir))
185+
# Extracting batch ID and hash from preexisting batch directory
186+
compiled = [i.split(".")[0].split("_") for i in sorted(os.listdir(bch_dir))]
187+
compiled_hsh = {
188+
int(i[1]): i[2] for i in compiled
189+
}
190+
# Generating each batch
191+
for bch_id, bch_start in enumerate(range(0, len(posts), bch_size)):
192+
# For remaining files at the end not reaching the size of a batch
193+
if bch_start + bch_size > len(posts): break
194+
# Current range of indices and hash of date strings
195+
bch_range = list(range(bch_start, bch_start + bch_size))
196+
# Join contents of each post to generate hash
197+
hsh = hash_str("".join([
198+
File(f"{src_dir}{posts[i]}/index.md").read() for i in bch_range
199+
]))[-6:]
200+
existing_hsh = compiled_hsh.get(bch_id)
201+
# If this batch is already present
202+
if existing_hsh:
203+
if existing_hsh == hsh:
204+
print(f" Skipping batch: {bch_id} #{hsh}")
205+
continue
206+
else:
207+
print(f" Removing obsolete: {bch_id} #{existing_hsh} -> #{hsh}")
208+
os.remove(f"{bch_dir}compilation_{bch_id}_{existing_hsh}.pdf")
209+
print(f" Processing batch: {bch_id} #{hsh}")
210+
filename = f"compilation_{bch_id}_{hsh}"
211+
# Writing index.tex to be compiled
212+
File(f"{tmp_dir}index.tex").writelines([
213+
# Title
214+
"\\mlytitle{" + f"c13n \\#{bch_id}" + "}",
215+
*[
216+
# Dump file contents from each post directory
217+
File(f"{pbl_dir}{posts[i]}/index.tex").read()
218+
for i in bch_range
219+
]
220+
])
221+
# Compiling and cleaning
222+
texcomp("drvmly.ltx")
223+
shutil.copy(tmp_dir + "index.pdf", bch_dir + filename.lower() + ".pdf")
224+
shutil.rmtree(tmp_dir)
169225

170226
if len(sys.argv) != 2:
171-
print("make: Incorrect options...")
227+
print("make: Incorrect options...")
172228
elif sys.argv[1] == "post":
173-
post()
229+
post()
174230
elif sys.argv[1] == "batch":
175-
batch()
231+
batch()
176232
else:
177-
print("make: Doing nothing...")
233+
print("make: Doing nothing...")

0 commit comments

Comments
 (0)