#! py -3 """ Extract code examples from TIJ Director's Cut plain text file. Creates Ant build.xml file for each subdirectory. """ from pathlib import Path import sys, os import re import shutil import pprint import difflib from collections import defaultdict from betools import CmdLine, visitDir, ruler, head maindef = re.compile("public\s+static\s+void\s+main") # Leave this alone, relative path necessary: destination = Path('.') / "ExtractedExamples" sourceText = Path('.') / "TIJDirectorsCut.txt" github = Path(r'C:\Users\Bruce\Documents\GitHub\TIJ-Directors-Cut') examples = Path(r"C:\Users\Bruce\Dropbox\__TIJ4-ebook\ExtractedExamples") startBuild = """\ """ endBuild = """\ """ def extractExamples(): print("Extracting examples ...") if not destination.exists(): destination.mkdir() if not sourceText.exists(): print("Cannot find", sourceText) sys.exit() with sourceText.open("rb") as book: text = book.read().decode("utf-8", "ignore") for listing in (re.findall("^//:.*?///:~", text, re.DOTALL | re.MULTILINE) + re.findall("^#:.*?#:~", text, re.DOTALL | re.MULTILINE)): title = listing.splitlines()[0] if "//: as a special marker" in title: continue title = title.split()[1] # print(title) target = destination / Path(title) if not target.parent.exists(): target.parent.mkdir(parents=True) if "//:!" in listing: listing = "".join(listing.splitlines(keepends=True)[1:-1]) with target.open("w", newline='') as codeListing: codeListing.write(listing) codeListing.write("\n") def clean(): "Remove ExtractedExamples directory" print("Cleaning ...") if destination.exists(): shutil.rmtree(str(destination)) @CmdLine("c") def compareWithGithub(shortForm=True): "Compare files from Github repository to extracted examples" leader = len(str(github)) + 1 githubfiles = [str(file)[leader:] for file in github.glob("**/*")] githubfiles = [ghf for ghf in githubfiles if not ghf.startswith(".git")] duplicates = { ghf for ghf in githubfiles if githubfiles.count(ghf) > 1 } if duplicates: print("duplicates = ", duplicates) leader2 = len(str(destination)) + 1 destfiles = [str(file)[leader2:] for file in destination.glob("**/*")] duplicates = { ghf for ghf in destfiles if destfiles.count(ghf) > 1 } if duplicates: print("duplicates = ", duplicates) githubfiles = set(githubfiles) destfiles = set(destfiles) runOutput = re.compile("/\* Output:.*///:~", re.DOTALL) differ = difflib.Differ() def rstrip(lines): return [line.rstrip() for line in lines] def show(lines, sep="#"): sys.stdout.writelines(lines) print("\n" + sep * 80) inBoth = [f for f in destfiles.intersection(githubfiles) if f.endswith(".java")] for f in inBoth: with (github / f).open() as ghf: with (destination / f).open() as dstf: ghblock = runOutput.sub("", ghf.read()) dstblock = runOutput.sub("", dstf.read()) if ghblock.strip() == dstblock.strip(): continue ghtext = ghblock.splitlines(keepends=True) dsttext = dstblock.splitlines(keepends=True) print("[[[", f, "]]]") if shortForm: show([ln + "\n" for ln in difflib.context_diff(rstrip(ghtext), rstrip(dsttext))], sep="=") else: show([ln + "\n" for ln in differ.compare(rstrip(ghtext), rstrip(dsttext))], sep="=") # def githubDirs(): # leader = len(str(github)) + 1 # buildfiles = [str(file)[leader:] for file in github.glob("**/build.xml")] # return {str((github / f).parent)[leader:] for f in buildfiles} # def destDirs(pattern="**"): # leader = len(str(destination)) + 1 # return {str(file)[leader:] for file in destination.glob(pattern)} def copySupplementalFilesFromGithub(): "Copy supplemental files from Github repository to extracted examples" print("Copying supplemental files from Github ...") def _copy_from_github(dir, name_or_pattern, trace=False): source = (github/dir).glob(name_or_pattern) dest_dir = examples/dir assert dest_dir.is_dir() for f in source: if trace: print("source: {}".format(f)) print("dest: {}".format(dest_dir)) shutil.copy(str(f), str(dest_dir)) for args in [ (".", "build.xml"), (".", "Ant-*.xml"), ("gui", "*.gif"), ("network", "*.bat"), ("network", "build.xml"), ("remote", "*.bat"), ("remote", "build.xml"), ]: _copy_from_github(*args) patterns = destination / "patterns" trash = patterns / "recycleap" / "Trash.dat" shutil.copy(str(trash), str(patterns / "recycleb")) shutil.copy(str(trash), str(patterns / "dynatrash")) class CodeFileOptions(object): """docstring for CodeFileOptions""" def __init__(self, codeFile): "Should probably use regular expressions for parsing instead" self.codeFile = codeFile self.msg = "" self.cmdargs = None if "{Args:" in self.codeFile.code: for line in self.codeFile.lines: if "{Args:" in line: self.cmdargs = line.split("{Args:")[1].strip() self.cmdargs = self.cmdargs.rsplit("}", 1)[0] self.validatebyhand = "{ValidateByHand}" in self.codeFile.code self.exclude = None if "{CompileTimeError}" in self.codeFile.code: self.exclude = self.codeFile.name + ".java" if self.codeFile.subdirs: self.exclude = '/'.join(self.codeFile.subdirs) + '/' + self.exclude self.continue_on_error = None if "{ThrowsException}" in self.codeFile.code: self.continue_on_error = True self.msg = "* Exception was Expected *" self.alternatemainclass = None if "{main: " in self.codeFile.code: for line in self.codeFile.lines: if "{main:" in line: self.alternatemainclass = line.split("{main:")[1].strip() self.alternatemainclass = self.alternatemainclass.rsplit("}", 1)[0] self.timeout = None if "{TimeOut:" in self.codeFile.code: for line in self.codeFile.lines: if "{TimeOut:" in line: self.timeout = line.split("{TimeOut:")[1].strip() self.timeout = self.timeout.rsplit("}", 1)[0] self.continue_on_error = True elif "//: gui/" in self.codeFile.code or "//: swt/" in self.codeFile.code or "{TimeOutDuringTesting}" in self.codeFile.code: self.timeout = "4000" self.continue_on_error = True self.msg = "* Timeout for Testing *" def classFile(self): start = """ \n" class CodeFile: def __init__(self, javaFile, chapterDir): self.chapter_dir = chapterDir self.java_file = javaFile self.subdirs = str(javaFile.parent).split("\\")[2:] with javaFile.open() as j: self.code = j.read() self.lines = self.code.splitlines() self.main = None if maindef.search(self.code): self.main = True self.package = None if "package " in self.code: for line in self.lines: if line.startswith("package ") and line.strip().endswith(";"): self.package = line break self.tagLine = self.lines[0][4:] self.relpath = '../' + '/'.join(self.tagLine.split('/')[:-1]) self.name = javaFile.name.split('.')[0] self.options = CodeFileOptions(self) def run_command(self): if not self.main: return "" return self.options.createRunCommand() def __repr__(self): result = self.tagLine if self.package: result += "\n" + self.package result += "\n" return result def packageName(self): return self.package.split()[1][:-1] def checkPackage(self): if not self.package: return True path = '.'.join(self.tagLine.split('/')[:-1]) packagePath = self.packageName() return path == packagePath class Chapter: def __init__(self, dir): self.dir = dir self.code_files = [CodeFile(javaFile, dir) for javaFile in dir.glob("**/*.java")] self.excludes = [cf.options.exclude for cf in self.code_files if cf.options.exclude] def __repr__(self): result = "-" * 80 result += "\n" + str(self.dir) + "\n" result += "-" * 80 result += "\n" for cf in self.code_files: result += str(cf.name) + "\n" return result def checkPackages(self): for cf in self.code_files: if not cf.checkPackage(): print("BAD PACKAGE") print("\t", cf.tagLine) print("\t", cf.package) print("\n".join(cf.lines)) def makeBuildFile(self): buildFile = startBuild % (self.dir.name, " ".join(self.excludes)) for cf in self.code_files: if any([cf.name + ".java" in f for f in self.excludes]) or cf.options.validatebyhand: # print("Excluding {}".format(cf)) continue buildFile += cf.run_command() buildFile += endBuild with (self.dir / "build.xml").open("w") as buildxml: buildxml.write(buildFile) exec = """\ """ #@CmdLine("m") def createAntFiles(): "Make ant files" print("Creating Ant Files ...") chapters = [Chapter(fd) for fd in destination.glob("*") if fd.is_dir() if not (fd / "build.xml").exists()] for chapter in chapters: chapter.checkPackages() chapter.makeBuildFile() @CmdLine("f") def findNonJavaFiles(): "Find non-java files in TIJDirectorsCut.txt" if not sourceText.exists(): print("Cannot find", sourceText) sys.exit() with sourceText.open("rb") as book: text = book.read().decode("utf-8", "ignore") for listing in re.findall("^//:.*?///:~", text, re.DOTALL | re.MULTILINE): title = listing.splitlines()[0].strip() if not title.endswith(".java"): print(title) @CmdLine('e') def extractAndCreateBuildFiles(): "Clean, then extract examples from TIJDirectorsCut.txt, build ant files" clean() extractExamples() createAntFiles() copySupplementalFilesFromGithub() os.chdir("ExtractedExamples") with open("run.bat", 'w') as run: run.write(r"python ..\Validate.py -p" + "\n") run.write(r"powershell .\runall.ps1" + "\n") run.write(r"python ..\Validate.py -e" + "\n") def generateAntClean(): "Generate directives for Ant-Clean.xml" others = set([f.name for f in examples.rglob("*") if not f.is_dir() if not f.suffix == ".java" if not f.suffix == ".class" if not f.suffix == ".py" if not f.suffix == ".cpp" if not str(f).endswith("-output.txt") if not str(f).endswith("-erroroutput.txt") if f.name ]) for f in others: print(""" """.format(f)) def findTags(lines): tagRE = re.compile("{.*?}", re.DOTALL) topblock = [] for line in lines: if line.startswith("//"): topblock.append(line) else: break topblock = [line[2:].strip() for line in topblock] tags = tagRE.findall(" ".join(topblock)) return tags @CmdLine('t') def findAllCommentTags(): "Find all '{}' comment tags in Java files" tagdict = defaultdict(list) for jf in [f for f in examples.rglob("*.java")]: with jf.open() as code: lines = code.readlines() tags = findTags(lines) if tags: # head(jf.name) # print("\n".join(tags)) for t in tags: tagdict[t].append(jf.name) pprint.pprint(tagdict) if __name__ == '__main__': print(__doc__) CmdLine.run()