xinyin025/hello-algo
1
0
Fork 0
mirror of https://github.com/krahets/hello-algo.git synced 2026-07-12 23:36:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix Repo

Browse Source
This commit is contained in:
krahets
2023-02-08 04:17:26 +08:00
parent 7f4efa6d5e
commit 0407cc720c
347 changed files with 150 additions and 132904 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/utils/build_markdown.py
+7 -1
View File
@@ -12,6 +12,7 @@ import glob
import shutil
from docs.utils.extract_code_python import ExtractCodeBlocksPython
from docs.utils.extract_code_java import ExtractCodeBlocksJava
from docs.utils.extract_code_cpp import ExtractCodeBlocksCpp
def build_markdown(md_path):
@@ -22,10 +23,12 @@ def build_markdown(md_path):
file_pattern = re.compile(r'\s*```(\w+)\s+title="(.+)"')
src_pattern = re.compile(r'\s*\[class\]\{(.*?)\}-\[func\]\{(.*?)\}')
for i in range(len(lines)):
i = 0
while i < len(lines):
# Find the line target to the source codes
src_match = src_pattern.match(lines[i])
if src_match is None:
i += 1
continue
for j in range(i - 1, -1 ,-1):
file_match = file_pattern.match(lines[j])
@@ -70,6 +73,8 @@ def build_markdown(md_path):
func_block = class_dict["funcs"][func_label]
for code_line in func_block["block"][::-1]:
lines.insert(header_line, code_line)
i += 1
with open(md_path.replace("docs/", "build/"), "w") as f:
f.writelines(lines)
@@ -79,6 +84,7 @@ def build_markdown(md_path):
extractor_dict = {
"java": ExtractCodeBlocksJava(),
"python": ExtractCodeBlocksPython(),
"cpp": ExtractCodeBlocksCpp(),
}
docs/utils/extract_code_cpp.py
+28
View File
@@ -0,0 +1,28 @@
"""
File: extract_code_python.py
Created Time: 2023-02-07
Author: Krahets (krahets@163.com)
"""
import re
import glob
import sys, os.path as osp
sys.path.append(osp.dirname(osp.dirname(osp.dirname(osp.abspath(__file__)))))
from docs.utils.extract_code_java import ExtractCodeBlocksJava
class ExtractCodeBlocksCpp(ExtractCodeBlocksJava):
def __init__(self) -> None:
super().__init__()
# Pattern to match function names and class names
self.func_pattern = r'(\s*)(static|)\s*(|\S+)\s*(\w+)(\(.*\))\s+\{'
self.class_pattern = r'(public|)\s*(class|struct)\s+(\w+)\s*\{'
self.func_pattern_keys = ["total", "ind", "static", "return", "label", "args"]
self.class_pattern_keys = ["total", "scope", "type", "label"]
# for code_path in glob.glob("codes/cpp/chapter_*/my_heap.cpp"):
# ext = ExtractCodeBlocksCpp()
# ext.extract(code_path)
docs/utils/extract_code_java.py
+20 -12
View File
@@ -5,19 +5,27 @@ Author: Krahets (krahets@163.com)
"""
import re
import os
import os.path as osp
import glob
import sys, os.path as osp
sys.path.append(osp.dirname(osp.dirname(osp.dirname(osp.abspath(__file__)))))
class ExtractCodeBlocksJava:
def __init__(self) -> None:
self.langs = ["java"]
self.ind = 4
# Pattern to match function names and class names
self.func_pattern = r'(\s+)(public|private|)\s*(static|)\s*(\S+)\s+(\w+)(\(.*\))\s+\{'
self.func_pattern = r'(\s*)(public|private|)\s*(static|)\s*(\S+)\s+(\w+)(\(.*\))\s+\{'
self.class_pattern = r'(public|)\s*class\s+(\w+)\s*\{'
self.func_pattern_keys = ["total", "ind", "scope", "static", "return", "label", "args"]
self.class_pattern_keys = ["total", "scope", "label"]
# Pattern to match the start and end of a block
self.block_end_pattern = '^\s{ind}\}'
self.block_start_pattern = '^\s{ind}\/\*.+\*\/'
self.block_start_shift = 0
self.block_end_shift = 0
def extract(self, file_path):
"""
@@ -53,17 +61,17 @@ class ExtractCodeBlocksJava:
# Search the code
for i in range(header_line + 1, len(self.lines)):
if re.match(block_end_pattern, self.lines[i]) is not None:
end_line = i
end_line = i + self.block_end_shift
break
# Search the header comment
for i in range(header_line - 1, -1, -1):
if re.search(block_start_pattern, self.lines[i]) is not None:
start_line = i
start_line = i + self.block_start_shift
break
return start_line, end_line, self.lines[start_line:end_line + 1]
def extract_function_blocks(self, indentation=4, start_line=-1, end_line=-1):
def extract_function_blocks(self, indentation=0, start_line=-1, end_line=-1):
"""
Extract all the functions with given indentation
"""
@@ -82,7 +90,7 @@ class ExtractCodeBlocksJava:
if func_match is None:
continue
# The function should match the input indentation
if len(func_match.group(1)) != indentation:
if len(func_match.group(self.func_pattern_keys.index("ind"))) != indentation:
continue
header_line = line_num
@@ -90,7 +98,7 @@ class ExtractCodeBlocksJava:
start_line, end_line, func_block = self.search_block(
header_line, indentation)
# Construct the funcs dict
func_label = func_match.group(5)
func_label = func_match.group(self.func_pattern_keys.index("label"))
funcs[func_label] = {
"indentation": indentation,
"line_number": {
@@ -122,7 +130,7 @@ class ExtractCodeBlocksJava:
start_line, end_line, class_block = self.search_block(
header_line, 0)
# Construct the classes dict
class_label = class_match.group(2)
class_label = class_match.group(self.class_pattern_keys.index("label"))
classes[class_label] = {
"indentation": 0,
"line_number": {
@@ -132,7 +140,7 @@ class ExtractCodeBlocksJava:
},
"block": class_block,
"funcs": self.extract_function_blocks(
indentation=4, start_line=start_line, end_line=end_line)
indentation=self.ind, start_line=start_line, end_line=end_line)
}
return classes
@@ -144,7 +152,7 @@ class ExtractCodeBlocksJava:
def remove_keyword(func):
block = func["block"]
header_line = func["line_number"]["header"] - \
func["line_number"]["start"]
func["line_number"]["start"]
block[header_line] = block[header_line] \
.replace("static ", "").replace("public ", "").replace("private ", "")
for clas in classes.values():
docs/utils/extract_code_python.py
+15 -117
View File
@@ -5,130 +5,28 @@ Author: Krahets (krahets@163.com)
"""
import re
import os
import os.path as osp
import glob
import sys, os.path as osp
sys.path.append(osp.dirname(osp.dirname(osp.dirname(osp.abspath(__file__)))))
from docs.utils.extract_code_java import ExtractCodeBlocksJava
class ExtractCodeBlocksPython:
class ExtractCodeBlocksPython(ExtractCodeBlocksJava):
def __init__(self) -> None:
self.langs = ["python"]
super().__init__()
# Pattern to match function names and class names
self.func_pattern = r'(\s*)def\s+(\w+)\s*\('
self.class_pattern = r'class\s+(\w+)'
self.func_pattern_keys = ["total", "ind", "label"]
self.class_pattern_keys = ["total", "label"]
# Pattern to match the start and end of a block
self.block_end_pattern = '^\s{0,ind}\S+.*\n'
self.block_start_pattern = '^\s{ind}""".+'
def extract(self, file_path):
"""
Extract classes and functions from a markdown document
"""
self.file_path = file_path
with open(file_path) as f:
self.lines = f.readlines()
self.content = "".join(self.lines)
# Detect and extract all the classes and fucntions
classes = self.extract_class_blocks()
funcs = self.extract_function_blocks()
self.post_process(classes, funcs)
return {
"classes": classes,
"funcs": funcs,
}
def search_block(self, header_line, indentation):
"""
Search class/function block given the header_line and indentation
"""
start_line, end_line = 0, len(self.lines)
block_end_pattern = re.compile(self.block_end_pattern.replace("ind", str(indentation)))
block_start_pattern = re.compile(self.block_start_pattern.replace("ind", str(indentation)))
# Search the code
for i in range(header_line + 1, len(self.lines)):
if re.match(block_end_pattern, self.lines[i]) is not None:
end_line = i - 1
break
# Search the header comment
for i in range(header_line - 1, -1, -1):
if re.search(block_start_pattern, self.lines[i]) is not None:
start_line = i
break
return start_line, end_line, self.lines[start_line:end_line + 1]
def extract_function_blocks(self, indentation=0, start_line=-1, end_line=-1):
"""
Extract all the functions with given indentation
"""
funcs = {}
if start_line == -1:
start_line = 0
if end_line == -1:
end_line = len(self.lines) - 1
func_pattern = re.compile(self.func_pattern)
for line_num in range(start_line, end_line + 1):
# Search the function header
func_match = func_pattern.match(self.lines[line_num])
if func_match is None: continue
# The function should match the input indentation
if len(func_match.group(1)) != indentation: continue
header_line = line_num
# Search the block from the header line
start_line, end_line, func_block = self.search_block(header_line, indentation)
# Construct the funcs dict
func_label = func_match.group(2)
funcs[func_label] = {
"indentation": indentation,
"line_number": {
"start": start_line,
"end": end_line,
"header": header_line,
},
"block": func_block,
}
return funcs
def extract_class_blocks(self):
"""
Extract all the classes with given indentation
"""
classes = {}
class_pattern = re.compile(self.class_pattern)
for line_num, line in enumerate(self.lines):
# Search the class header
class_match = class_pattern.match(line)
if class_match is None: continue
header_line = line_num
# Search the block from the header line
start_line, end_line, class_block = self.search_block(header_line, 0)
# Construct the classes dict
class_label = class_match.group(1)
classes[class_label] = {
"indentation": 0,
"line_number": {
"start": start_line,
"end": end_line,
"header": header_line,
},
"block": class_block,
"funcs": self.extract_function_blocks(
indentation=4, start_line=start_line, end_line=end_line)
}
return classes
self.block_start_shift = 0
self.block_end_shift = -1
def post_process(self, classes, funcs):
"""
@@ -154,4 +52,4 @@ class ExtractCodeBlocksPython:
# ext = ExtractCodeBlocksPython()
# ext.extract("codes/python/chapter_array_and_linkedlist/my_list.py")
# ext.extract("codes/python/chapter_array_and_linkedlist/my_list.py")