Speed up SVG generation and images link replacement, plus other optim…

…izations

mkdocs_juvix/common/models/wikilink.py

@@ -1,7 +1,9 @@
 from typing import Optional
 
-from .loc import FileLoc
 from mkdocs.plugins import get_plugin_logger
+
+from .loc import FileLoc
+
 log = get_plugin_logger("\033[94m[wikilinks]\033[0m")
 
 

mkdocs_juvix/common/preprocesors/links.py

@@ -1,16 +1,17 @@
 import os
 import re
-from pathlib import Path
 import time
+from pathlib import Path
 from typing import Any, Dict, List, Optional
 from urllib.parse import urljoin
 
-from mkdocs.plugins import get_plugin_logger
+import numpy as np  # type: ignore
+from colorama import Fore, Style  # type: ignore
 from fuzzywuzzy import fuzz  # type: ignore
 from markdown.preprocessors import Preprocessor  # type: ignore
+from mkdocs.plugins import get_plugin_logger
 from mkdocs.structure.pages import Page
-from ncls import NCLS # type: ignore
-import numpy as np # type: ignore
+from ncls import NCLS  # type: ignore
 
 from mkdocs_juvix.common.models import FileLoc, WikiLink
 from mkdocs_juvix.env import ENV
@@ -27,7 +28,8 @@
     re.VERBOSE,
 )
 
-log = get_plugin_logger("\033[94m[wikilinks_preprocessor]\033[0m")
+
+log = get_plugin_logger(f"{Fore.BLUE}[juvix_mkdocs-wikilinks]{Style.RESET_ALL}")
 
 REPORT_BROKEN_WIKILINKS = bool(os.environ.get("REPORT_BROKEN_WIKILINKS", False))
 
@@ -94,28 +96,32 @@ def run(self, lines) -> List[str]:
             cache_filepath
             and cache_filepath.exists()
             and original_filepath
-            and not self.env.new_or_changed_or_no_exist(original_filepath)
+            and not self.env.new_or_changed_or_not_exists(original_filepath)
         ):
             return cache_filepath.read_text().split("\n")
 
         if self.run_snippet_preprocessor:
             time_start = time.time()
             lines = self.snippet_preprocessor.run(lines)
             time_end = time.time()
-            log.info(f"Snippet finished in {(time_end - time_start):.5f} seconds")
+            log.info(
+                f"Snippet finished in {Fore.GREEN}{(time_end - time_start):.5f}{Style.RESET_ALL} seconds"
+            )
 
-        log.info(f"Processing wikilinks on file {filepath}")
+        log.info(
+            f"Processing wikilinks on file {Fore.GREEN}{filepath}{Style.RESET_ALL}"
+        )
         # Combine all lines into a single string
         full_text = "\n".join(lines)
         # Find all code blocks, HTML comments, and script tags in a single pass
-        pattern = re.compile(
+        ignore_blocks = re.compile(
             r"(```(?:[\s\S]*?)```|<!--[\s\S]*?-->|<script>[\s\S]*?</script>)", re.DOTALL
         )
 
         intervals = []
         time_start = time.time()
         try:
-            for match in pattern.finditer(full_text):
+            for match in ignore_blocks.finditer(full_text):
                 intervals.append((match.start(), match.end(), 1))
         except TimeoutError:
             log.error("Timeout occurred while processing ignore patterns")
@@ -150,8 +156,10 @@ def run(self, lines) -> List[str]:
         for start, end, new_text in reversed(replacements):
             full_text = full_text[:start] + new_text + full_text[end:]
         time_end = time.time()
-
-        log.info(f"Processing wikilinks took {(time_end - time_start):.5f} seconds")
+
+        log.info(
+            f"Processing wikilinks took {Fore.GREEN}{(time_end - time_start):.5f}{Style.RESET_ALL} seconds"
+        )
 
         if cache_filepath:
             log.debug(f"Writing wikilinks to cache for file {original_filepath}")

mkdocs_juvix/common/utils.py

@@ -18,12 +18,15 @@ def fix_site_url(config: MkDocsConfig) -> MkDocsConfig:
             config["site_url"] += "/"
         return config
 
-    log.info("SITE_URL environment variable not set")
+    log.debug("SITE_URL environment variable not set")
 
     version = os.environ.get("MIKE_DOCS_VERSION")
 
     if version:
-        log.info(f"Using MIKE_DOCS_VERSION environment variable: {version}")
+        log.debug(f"Using MIKE_DOCS_VERSION environment variable: {version}")
+
+    if "site_url" not in config or not config["site_url"]:
+        config["site_url"] = ""
 
     if not config["site_url"].endswith("/"):
         config["site_url"] += "/"

mkdocs_juvix/env.py

@@ -4,26 +4,28 @@
 Juvix settings.
 """
 
-from functools import lru_cache
 import os
 import shutil
 import subprocess
+from functools import lru_cache
 from os import getenv
 from pathlib import Path
 from typing import List, Optional, Tuple
 
+from colorama import Fore, Style  # type: ignore
 from mkdocs.config.defaults import MkDocsConfig
 from mkdocs.plugins import get_plugin_logger
 from semver import Version
 
 from mkdocs_juvix.juvix_version import MIN_JUVIX_VERSION
 from mkdocs_juvix.utils import compute_hash_filepath, hash_file
 
-log = get_plugin_logger("ENV")
+log = get_plugin_logger(f"{Fore.BLUE}[juvix_mkdocs-env]{Style.RESET_ALL}")
 
 BASE_PATH = Path(__file__).parent
 FIXTURES_PATH = BASE_PATH / "fixtures"
 
+
 class ENV:
     ROOT_PATH: Path
     DOCS_DIRNAME: str = getenv("DOCS_DIRNAME", "docs")
@@ -94,7 +96,6 @@ class ENV:
         "CACHE_WIKILINKS_DIRNAME", ".wikilinks"
     )  # The name of the file where the Juvix Markdown files are stored
     DOCS_IMAGES_DIRNAME: str = getenv("DOCS_IMAGES_DIRNAME", "images")
-    CACHE_IMAGES_DIRNAME: str = getenv("CACHE_IMAGES_DIRNAME", ".images")
     CACHE_JUVIX_VERSION_FILENAME: str = getenv(
         "CACHE_JUVIX_VERSION_FILENAME", ".juvix_version"
     )  # The name of the file where the Juvix version is stored
@@ -120,7 +121,6 @@ class ENV:
     SHOW_TODOS_IN_MD: bool
     INDEXES_PATH: Path
     IMAGES_PATH: Path
-    CACHE_IMAGES_PATH: Path
 
     def __init__(self, config: Optional[MkDocsConfig] = None):
         if config:
@@ -196,9 +196,6 @@ def __init__(self, config: Optional[MkDocsConfig] = None):
         self.CACHE_HASHES_PATH: Path = (
             self.CACHE_ABSPATH / self.CACHE_HASHES_DIRNAME
         )  # The path where hashes are stored (not the project hash)
-        self.CACHE_IMAGES_PATH: Path = (
-            self.CACHE_ABSPATH / self.CACHE_IMAGES_DIRNAME
-        )  # The path to the images cache directory
 
         self.JUVIX_FOOTER_CSS_FILEPATH: Path = (
             self.DOCS_ABSPATH / "assets" / "css" / self.JUVIX_FOOTER_CSS_FILENAME
@@ -224,12 +221,19 @@ def __init__(self, config: Optional[MkDocsConfig] = None):
             self.CACHE_HASHES_PATH,
             self.JUVIX_FOOTER_CSS_FILEPATH.parent,
             self.CACHE_WIKILINKS_PATH,
-            self.CACHE_IMAGES_PATH,
         ]
 
         for directory in directories:
-            if directory.exists() and self.REMOVE_CACHE:
+            if (
+                directory.exists()
+                and self.REMOVE_CACHE
+                and config
+                and not config.get("env_init", False)
+            ):
                 try:
+                    log.info(
+                        f"Removing directory {Fore.RED}{directory}{Style.RESET_ALL}"
+                    )
                     shutil.rmtree(directory, ignore_errors=True)
                 except Exception as e:
                     log.error(
@@ -281,11 +285,13 @@ def __init__(self, config: Optional[MkDocsConfig] = None):
             self.JUVIX_ENABLED = False
             self.JUVIX_AVAILABLE = False
             return
-    
+
         self.USE_DOT = bool(getenv("USE_DOT", True))
         self.DOT_BIN = getenv("DOT_BIN", "dot")
         self.DOT_FLAGS = getenv("DOT_FLAGS", "-Tsvg")
         self.IMAGES_ENABLED = bool(getenv("IMAGES_ENABLED", True))
+        if config:
+            config["env_init"] = True
 
     @lru_cache(maxsize=128)
     def read_markdown_file_from_cache(self, filepath: Path) -> Optional[str]:
@@ -298,37 +304,33 @@ def read_wikilinks_file_from_cache(self, filepath: Path) -> Optional[str]:
         if cache_ABSpath := self.get_filepath_for_wikilinks_in_cache(filepath):
             return cache_ABSpath.read_text()
         return None
-    
+
     def write_wikilinks_file_to_cache(self, filepath: Path, content: str) -> None:
         if cache_ABSpath := self.get_filepath_for_wikilinks_in_cache(filepath):
             cache_ABSpath.write_text(content)
-    
+
     def get_filepath_for_wikilinks_in_cache(self, filepath: Path) -> Optional[Path]:
         filepath = filepath.absolute()
         rel_to_docs = filepath.relative_to(self.DOCS_ABSPATH)
-        return (
-            self.CACHE_WIKILINKS_PATH / rel_to_docs.parent / filepath.name
-        )
+        return self.CACHE_WIKILINKS_PATH / rel_to_docs.parent / filepath.name
 
-    def new_or_changed_or_no_exist(self, filepath: Path) -> bool:
+    def new_or_changed_or_not_exists(self, filepath: Path) -> bool:
         content_hash = hash_file(filepath)
         path_hash = compute_hash_filepath(filepath, hash_dir=self.CACHE_HASHES_PATH)
         if not path_hash.exists():
             log.debug(f"File: {filepath} does not have a hash file.")
             return True
         fresh_content_hash = path_hash.read_text()
         return content_hash != fresh_content_hash
-    
+
     @lru_cache(maxsize=128)
     def get_filepath_for_juvix_markdown_in_cache(
         self, _filepath: Path
     ) -> Optional[Path]:
         filepath = _filepath.absolute()
         md_filename = filepath.name.replace(".juvix.md", ".md")
         rel_to_docs = filepath.relative_to(self.DOCS_ABSPATH)
-        return (
-            self.CACHE_MARKDOWN_JUVIX_OUTPUT_PATH / rel_to_docs.parent / md_filename
-        )
+        return self.CACHE_MARKDOWN_JUVIX_OUTPUT_PATH / rel_to_docs.parent / md_filename
 
     def unqualified_module_name(self, filepath: Path) -> Optional[str]:
         fposix: str = filepath.as_posix()
@@ -371,9 +373,7 @@ def get_filename_module_by_extension(
         return module_name + extension if module_name else None
 
     def update_hash_file(self, filepath: Path) -> Optional[Tuple[Path, str]]:  # noqa: F821
-        filepath_hash = compute_hash_filepath(
-            filepath, hash_dir=self.CACHE_HASHES_PATH
-        )
+        filepath_hash = compute_hash_filepath(filepath, hash_dir=self.CACHE_HASHES_PATH)
         try:
             with open(filepath_hash, "w") as f:
                 content_hash = hash_file(filepath)
@@ -398,4 +398,3 @@ def get_expected_filepath_for_juvix_isabelle_output_in_cache(
             / cache_isabelle_filename
         )
         return cache_isabelle_filepath
-