tags - jinja: add newline and/or empty lines depending on the Command type and remove empty lines from the jinja specs themselves or respectively add strip() to the initial parser of the jinja specs - Json serialization - XML output - EPUB output - Parsing for XML/HTML - argument --allformats that produces output in all formats plus json - PACKAGE DISTRIBUTION: pip installation and the like - consistency of tags, especially
, probably all misplaced par instances must be deleted ... - boxes (should already work but need to be tested, rendering needs to be optimized) - rotateboxes (especially in klammerabsatz macros) - XML rendering - Json output, mere text output - docx rendering - epub rendering - rendering: -- functions in expressions, e.g., to calculate correct lengths and the like -- bug fixes, consistent and valid document structure! - tables: -- cline commands (not yet interpreted at all) -- tabbing: add interpretation for esoteric tabstop commands -- cell width (probably with optional arguments ? ) -- rowspan and colspan - data types and functions in rendering expressions -- e.g. calculations of lengths -- klammerabsatz e.g. Mai 2-4 1913 Formats: multiple underlinings in different colors (probably mainly CSS and/or Javascript) - graphics with includegraphics - batch conversion of eps files to svg (is not working properly) - theorems later: - Formulas need to be optimized the rendering does not yet work well!! - def macro definitions - Static Table of Contents - Bibliographies ''' ################################################################################### # LOAD DEFAULT FILE OPENING: # this needs to be done right at the beginning, before calling parse_latex def load_default_preamble(stack): path = os.path.dirname(os.path.abspath(__file__)) + '/custom/default-fileopening.tex' if not os.path.exists(path): path = os.path.dirname(os.path.abspath(__file__)) + '/default-fileopening.tex' # Read the default file opening try: with open(path, 'r', encoding='utf-8') as f: default_fileopening = f.read() f.close() except: stack.error += f'\n ERROR: File "{path}" could not be opened for parsing!\n' print(f'ERROR: File "{path} could not be opened for parsing!') default_fileopening = '' stack = load_preamble_string(stack, default_fileopening) if stack.defaultpreamble == False: stack.packages.clear() return stack # END LOADING DEFAULT FILE OPENING #################################### ################################################################################ # the following three functions are needed for # LOADING ALL JINJA SPECIFICATIONS # = loads all files with the extension .jinja2s from directory def find_files(directory, pattern): file_paths = [] for root, _, files in os.walk(directory): for filename in files: if fnmatch.fnmatch(filename, pattern): file_paths.append(os.path.join(root, filename)) return file_paths folder_path = "path" # Replace "path" with your folder path file_paths = find_files(folder_path, "*.jinja2h") # this is currently not needed: def add_jinja_file(path): try: with open(f'{path}', 'r', encoding='utf-8') as f: string = f.read() return f'{string}\n' except: print(f'ERROR: File {path} could not be opened for parsing!') stack.error += f'\n ERROR: File {path} could not be opened for parsing!\n' return '' def load_jinja_specs(stack, path): jinja_paths = find_files(path, '*.jinja2s') string = '' for jinja_path in jinja_paths: string += add_jinja_file(jinja_path) if string: # ... and parse jinja file: stack.jinja = JinjaSpec(string, stack) else: print(f'ERROR: No jinja specifications could be found in path "{path}"!') return None return stack # END LOADING ALL JINJA SPECS ########################################## ################################################################# # LOAD JINJA TEMPLATE # first check the custom subfolder for a file default-layout.jinja2 # if this fails load the default-layout.jinja2 from the root folder def load_jinja_template(stack, path): path = path + '/default-layout.jinja2' if not os.path.exists(path): print('ERROR: no jinja template found!!') return None stack.template_string = add_jinja_file(path) return stack # END LOADING JINJA TEMPLATE ########################################## ########################################################################################## # Main function parse_latex: # - loads the tex file and packages # - calls all necessary parsing functions # - returns a html string # def parse_latex(tex_file_name, stack): if tex_file_name[-4:] == '.tex' or tex_file_name[-4:] == '.TEX': tex_file_name = tex_file_name[0:-4] tex_file_name = str(tex_file_name) stack.filename = tex_file_name print(f'This is valepTeX parser version 0.1 (Mai 2024): parsing of document {tex_file_name} begins:') # transient elements of stack need to be reset or emptied: stack.reset_counters() stack.reset_transient_elements() # FIRST STEP # start the parsing process stack.preamblemode = True tex_file, stack = start_parse(tex_file_name, stack) # SECOND AND MAIN STEP: # parse blocks doc_list = DocNode() doc_list, stack = parse_blocks(tex_file, stack) print('... optimizes DOM tree and renders document ...') # THIRD STEP: brush up the DOM tree # (1) deal with notes of the form \NOTEmark and \NOTEtext: doc_list.supplement_note_marks(stack) # (2) flatten the docnode to facilitate iterating over the docnode lists of the DOM tree: doc_list.elements = doc_list.flatten_docnodes() # (3) pick out the document part: match = doc_list.find_object(document) dom_object = DocNode() if match: dom_object.elements.append(match) else: print('WARNING: incorrect document structure!') dom_object = doc_list # (4) adjust paragraphs: dom_object.adjust_paragraphs() # FOURTH STEP: # render document string = dom_object.render(stack.jinja) # FIFTH STEP: # embed the rendered document in the page template template = JinjaDocument(string, stack.template_string, stack) string = dom_object.render_document(template, string) # SIXTH STEP: clean up # remove empty
tags string = regex.sub(r'
\s*<\/p>','', string) string = regex.sub(r'
\s*
',r'
', string) # this is just an ad hoc fix! string = regex.sub(r'<\/p>\s*<\/p>',r'
', string) # this is just an ad hoc fix! # deal with xml specialities: if stack.output_format == 'xml': html_entities = { "ü": "ü", "ä": "ä", "ö": "ö", "Ü": "Ü", "Ä": "Ä", "Ö": "Ö", "ß": "ß", "⁄": "⁄", "·": "·", # Fügen Sie hier weitere HTML-Sonderzeichen nach Bedarf hinzu } for entity, char in html_entities.items(): string = string.replace(entity, char) ########################################################## # THIS IS VERY AD HOC AND NEEDS TO BE GENERALIZED FOR OTHER VERBATIM Commands # if its html or xml: clean uptags to avoid gettinginterpreted as tags def repl(match): arg = match.group(1) arg = arg.replace('<','<') arg = arg.replace('>','>') replacement = f' {arg}' return replacement string = regex.sub(r'([\s\S]*?)', repl, string) if stack.normalize: normal_text = doc_list.normalize(stack.jinja).strip() try: with open(f'{tex_file_name}.txt', 'w', encoding='utf-8') as f: f.write(normal_text) except: print(f'ERROR: could not write {tex_file_name}.text!') # only carnap edition: if stack.carnapedition: ###################################################################### # IDIOSYNCRATIC STUFF CARNAP EDITION # POSTPROCESSING HTML FILE WITH # postprocess the html file with functions from htnmlcarnap # add counters and labels to notes, indices, and readinglists string = replace_tag_index(string, 'in-dex class="person"', '') string = replace_tag_index(string, 'in-dex class="institution"', '') string = replace_tag_index(string, 'in-dex class="location"', '') string = replace_tag_index(string, 'in-dex class="sache"', '') # needs to be extended with: concepts, other doc categories ... string = replace_tag_index(string, 'in-dex class="doc"', 'B') string = replace_tag_index(string, 'fac-simile', '') string = replace_tag_index(string, 'loc-ation', 'O') string = add_all_to_reading_lists(string) string = replace_triggers(string) # resolve references to diary entries string = resolve_diary_references(string) # add content to diary entry time tags and add monthly label in left margin string = fill_time_tags(string) # END CARNAP EDITION STUFF ###################################################################### # add title to headline and task bar and produce the table of content string, stack = make_title_toc(string, stack) # make bilingual representations using the format of the Carnap edition string, stack = merge_bilingual(string, stack) #print(normal_text) # SEVENTH STEP: # return string and stack: html_content = string # dirty trick for correct processing of \NOTEmark and \NOTEtext notes: stack.firstdoc = False return html_content, stack # # end main function parse_latex ####################################### # write the log and/or unknown commands in a logfile def write_log(stack): log = '' if stack.writeunknown: log = stack.expand_unknown() if stack.writelog: log = stack.expand() log += stack.error if log: path = os.path.dirname(os.path.abspath(__file__)) + '/valeptex.log' with open(path, 'w', encoding = 'utf-8') as f: f.write(log) f.close() ####################################### # calls parse_latex and # writes the returned html string # in a file tex_tile_name.html def parse_latex_store_html(tex_file_name, stack): # remove file ending .tex if applicable if tex_file_name[-4:] == '.tex': tex_file_name = tex_file_name[0:-4] # call main function parse_latex html_content, stack = parse_latex(tex_file_name, stack) # Write rendered file to a .html file temp_tex_file_name = f'{tex_file_name}.{stack.output_format}' try: with open(temp_tex_file_name, 'w', encoding='utf-8') as f: f.write(html_content) except: print(f'ERROR: could not write {temp_tex_file_name}!') undefined = stack.expand_unknown() #print(undefined) return stack ############################################################################################### # BATCH PROCESSING # batch processing of parse_latex: either convert all .tex files in the current folder or # in the current folder together with all its subfolders ::: options --folder and --subfolders def convert_folder_tex_to_html(folder_path, stack, include_subfolders=False): pattern = os.path.join(folder_path, '**/*.tex') if include_subfolders else os.path.join(folder_path, '*.tex') error = "" for tex_file in glob.iglob(pattern, recursive=True): # change to subdir directory = os.path.dirname(tex_file) os.chdir(directory) # use relativ filepath tex_file = os.path.relpath(tex_file) try: stack = parse_latex_store_html(tex_file, stack) except: print(f"ERROR: conversion of file {tex_file} failed.") error += f"ERROR: conversion of file {tex_file} failed." write_log(stack) # Function to convert EPS to SVG using matplotlib with improved quality def convert_eps_to_svg(input_path, output_path, dpi=300): try: # Load the EPS file using matplotlib fig = plt.figure(dpi=dpi) plt.axis('off') plt.imshow(plt.imread(input_path)) # Save the figure as SVG with improved quality plt.savefig(output_path, format='svg', bbox_inches='tight', pad_inches=0) plt.close(fig) print(f"Converted {input_path} to {output_path} with DPI={dpi}") except Exception as e: print(f"Error converting {input_path}: {e}") # Function to recursively search for EPS files and convert them to SVG def convert_eps_in_directory(directory): for root, _, files in os.walk(directory): for file in files: if file.lower().endswith('.eps'): input_path = os.path.join(root, file) output_path = os.path.splitext(input_path)[0] + '.svg' convert_eps_to_svg(input_path, output_path) if __name__ == "__main__": parser = argparse.ArgumentParser(description="Parse LaTeX") parser.add_argument("tex_file", nargs="?", help="Path to the .tex file to convert") parser.add_argument("--folder", action="store_true", help="Convert .tex files in the current folder") parser.add_argument("--subfolders", action="store_true", help="Convert .tex files in all subfolders") parser.add_argument("--normalize", action="store_true", help="save also a .txt version of the document") parser.add_argument("--nologfile", action="store_true", help="do not save the valeptex.log") parser.add_argument("--writeunknown", action="store_true", help="only write the unknown commands in the logfile") parser.add_argument("--useownpreamble", action="store_true", help="use the original preamble of the file") parser.add_argument("--convertsvg", action="store_true", help="use the original preamble of the file") parser.add_argument("--format", help="Specify the file format by its file ending") args = parser.parse_args() # ATTENTION: DocStack needs to be created and filled before calling parse_latex # this would be necessary also when calling parse_latex from somewhere else stack = DocStack() if args.format and args.format != 'html': file_format = args.format script_dir = os.path.dirname(os.path.abspath(__file__)) subfolder_path = os.path.join(script_dir, file_format) if os.path.exists(subfolder_path) and os.path.isdir(subfolder_path): stack.output_format = args.format path = subfolder_path stack = load_jinja_specs(stack, path) if stack == None: sys.exit(1) # load the jinja template stack = load_jinja_template(stack, path) if stack == None: sys.exit(1) print(f"Identified format folder {file_format} and loaded jinja specs.") else: print(f"WARNING: subfolder '{file_format}' does not exist! HTML rendering is chosen instead ...") stack.output_format = 'html' if stack.output_format == 'html': script_dir = os.path.dirname(os.path.abspath(__file__)) path = os.path.join(script_dir, 'html') stack = load_jinja_specs(stack, path) if stack == None: sys.exit(1) # load the jinja template stack = load_jinja_template(stack, path) if stack == None: sys.exit(1) print('Rendering information for html has been loaded.') stack = load_default_preamble(stack) # put all jinja specs in stack if args.normalize: stack.normalize = True if args.nologfile: stack.writelog = False if args.writeunknown: stack.writeunknown = True if args.useownpreamble: stack.defaultpreamble = False if args.tex_file: stack = parse_latex_store_html(args.tex_file, stack) write_log(stack) elif args.folder: current_folder = os.getcwd() print("Converting .tex files in the current folder...") convert_folder_tex_to_html(current_folder, stack) elif args.subfolders: current_folder = os.getcwd() print("Converting .tex files in all subfolders...") convert_folder_tex_to_html(current_folder, stack, include_subfolders=True) elif args.convertsvg: eps_directory = os.getcwd() #convert_eps_in_directory(eps_directory) inkscape_path = r"C:\Program Files\Inkscape\bin\inkscape.exe" for root, dirs, files in os.walk(eps_directory): for file in files: if file.endswith('.eps'): eps_file = os.path.join(root, file) svg_file = os.path.splitext(eps_file)[0] + '.svg' print(f"Converting {eps_file} to {svg_file}") subprocess.run([inkscape_path, '--export-plain-svg=' + svg_file, eps_file], stdout=subprocess.PIPE, stderr=subprocess.PIPE) else: print("Error: Please specify either a .tex file or --folder, --subfolders, or both.")