move processing to dedicated file

This commit is contained in:
Kylian Schmidt
2026-04-22 13:08:47 +02:00
parent a8f2166dde
commit 45214be41e
4 changed files with 316 additions and 272 deletions
+2
View File
@@ -43,5 +43,7 @@ sources:
path: "/work/kschmidt/NEEDLE/test_analysis/data/cf_store/test_analysis/cf.PlotVariables1D/run2_2016_nano_v9/"
- name: "needle_benchmarks"
path: "/work/kschmidt/NEEDLE/orchestrator/ml/benchmarks/plots"
- name: "needle_fair_universe"
path: "/ceph/kschmidt/needle/plots/fair_universe/"
- name: "aido_convergence_study"
path: "/work/kschmidt/aido/results_convergence/plots/"
+22 -272
View File
@@ -13,7 +13,6 @@ Features:
- Responsive grid layout with search and navigation
"""
import subprocess
import shutil
from pathlib import Path
from typing import Dict, Any, Optional
@@ -24,9 +23,12 @@ from utils.config import Config
from utils.metadata import (
load_folder_metadata,
merge_metadata,
resolve_metadata_for_plot,
save_metadata_cache,
get_metadata_file_path
)
from utils.processing import (
process_plot_files,
needs_update,
render_gallery_page,
)
@@ -49,213 +51,6 @@ env.filters['strftime'] = strftime_filter
template = env.get_template("templates/gallery.html")
def process_html_file(html_file: Path, web_dir: Path, current_metadata: Dict[str, Any] = None) -> dict:
"""
Process HTML plot file, copying it to web directory.
Args:
html_file: Path to the source HTML file
web_dir: Target web directory
current_metadata: Current metadata dictionary for the plot
Returns:
Dictionary containing plot information
"""
web_html = web_dir / html_file.name
if needs_update(html_file, web_html):
shutil.copy2(html_file, web_html)
else:
print(f"Skipping {html_file.name} (up to date)")
# Get source file creation time
source_creation_time = int(html_file.stat().st_ctime)
# Resolve metadata if provided
plot_metadata = {}
if current_metadata is not None:
plot_metadata = resolve_metadata_for_plot(html_file, current_metadata)
return {
"name": html_file.stem,
"html_href": html_file.name,
"is_html": True,
"metadata": plot_metadata,
"creation_time": source_creation_time
}
def process_plot_files(
plot_file: Path,
web_dir: Path,
current_metadata: Dict[str, Any] = None,
) -> dict:
"""
Process plot files (PDF/PNG or HTML), handling conversion and copying.
Args:
plot_file: Path to the source plot file (PDF or HTML)
web_dir: Target web directory
current_metadata: Current metadata dictionary for the plot
Returns:
Dictionary containing plot information
"""
if plot_file.suffix.lower() == '.html':
return process_html_file(plot_file, web_dir, current_metadata)
# Handle PDF files
png_file = plot_file.with_suffix(".png")
web_pdf = web_dir / plot_file.name
web_png = web_dir / png_file.name
if needs_update(plot_file, web_pdf):
shutil.copy2(plot_file, web_pdf)
else:
print(f"Skipping {plot_file.name} (up to date)")
if not png_file.exists():
convert_pdf_to_png(plot_file)
if needs_update(png_file, web_png):
shutil.copy2(png_file, web_png)
else:
print(f"Skipping {png_file.name} (up to date)")
source_creation_time = int(plot_file.stat().st_ctime)
plot_metadata = {}
if current_metadata is not None:
plot_metadata = resolve_metadata_for_plot(plot_file, current_metadata)
return {
"name": plot_file.stem,
"pdf_href": plot_file.name,
"png_href": png_file.name,
"is_html": False,
"metadata": plot_metadata,
"creation_time": source_creation_time
}
def render_gallery_page(
web_dir: Path,
items: list,
subdirs: list,
relative_path: Path,
title: str = None,
metadata: dict = None
) -> None:
"""
Unified template rendering for all gallery pages.
Args:
web_dir: Target web directory
items: List of plot items
subdirs: List of subdirectory names
relative_path: Relative path from gallery root
title: Page title (optional)
metadata: Metadata dictionary (optional)
"""
if title is None:
title = "Gallery" if relative_path == Path(
".") else f"Gallery: {relative_path}"
if metadata is None:
metadata = {}
# Calculate statistics
current_stats = calculate_directory_stats(web_dir)
stats = {
"file_count": len(items),
"folder_count": len(subdirs),
"total_size": format_file_size(current_stats["total_size"]),
"total_size_bytes": current_stats["total_size"]
}
# Calculate relative path to assets
if relative_path == Path("."):
assets_path = "../assets"
else:
depth = len(relative_path.parts)
assets_path = "../" * (depth + 1) + "assets"
# For root level, show only directory structure
if relative_path == Path("."):
items = []
output_html = web_dir / "index.html"
with output_html.open("w") as f:
rendered_html = template.render(
title=title,
items=items,
subdirs=subdirs,
relpath=str(relative_path),
paths=CONFIG.paths,
ui=CONFIG.ui,
stats=stats,
folder_metadata=metadata,
assets_path=assets_path,
source_dir=str(web_dir),
metadata_file_path=get_metadata_file_path(web_dir)
)
f.write(rendered_html)
print(f"Generated {output_html}")
def convert_pdf_to_png(pdf_path: Path) -> None:
"""
Convert a PDF file to PNG format using ImageMagick.
Only converts if the PNG doesn't exist or if the PDF is newer than
the PNG (with a 30-second buffer to handle filesystem timing issues).
Args:
pdf_path: Path to the source PDF file
Raises:
subprocess.CalledProcessError: If ImageMagick conversion fails
"""
png_path = pdf_path.with_suffix(".png")
if png_path.exists():
pdf_mtime = pdf_path.stat().st_mtime
png_mtime = png_path.stat().st_mtime
if png_mtime >= (pdf_mtime + 30):
return
else:
print(f"PDF {pdf_path.name} is newer than PNG, reconverting...")
print(f"Converting {pdf_path} to PNG")
subprocess.run([
"convert",
"-density", str(CONFIG.png_dpi),
str(pdf_path),
"-quality", "95",
str(png_path)
], check=True)
def needs_update(source_file: Path, target_file: Path) -> bool:
"""
Check if target file needs updating based on source modification time.
Args:
source_file: Path to the source file
target_file: Path to the target file
Returns:
True if target needs update, False otherwise
"""
if not target_file.exists():
return True
source_mtime = source_file.stat().st_mtime
target_mtime = target_file.stat().st_mtime
return source_mtime > (target_mtime + 30)
def build_gallery(
source_dir: Path,
web_dir: Path,
@@ -294,7 +89,12 @@ def build_gallery(
# Process all plot files (PDFs and HTMLs)
for plot_file in plot_files:
item = process_plot_files(plot_file, web_dir, current_metadata)
item = process_plot_files(
CONFIG=CONFIG,
plot_file=plot_file,
web_dir=web_dir,
current_metadata=current_metadata,
)
items.append(item)
plot_metadata_cache[plot_file.stem] = item["metadata"]
@@ -310,6 +110,8 @@ def build_gallery(
subdir_names.append(subdir.name)
render_gallery_page(
CONFIG=CONFIG,
template=template,
web_dir=web_dir,
items=items,
subdirs=subdir_names,
@@ -318,66 +120,6 @@ def build_gallery(
)
def calculate_directory_stats(directory: Path) -> dict:
"""
Calculate statistics for a directory.
Args:
directory: Path to the directory to analyze
Returns:
Dictionary containing file count, folder count, and total size
"""
stats = {
"file_count": 0,
"folder_count": 0,
"total_size": 0,
"pdf_size": 0,
"png_size": 0,
}
if not directory.exists():
return stats
for item in directory.rglob("*"):
if item.is_file():
stats["file_count"] += 1
size = item.stat().st_size
stats["total_size"] += size
if item.suffix.lower() == '.pdf':
stats["pdf_size"] += size
elif item.suffix.lower() == '.png':
stats["png_size"] += size
elif item.is_dir():
stats["folder_count"] += 1
return stats
def format_file_size(size_bytes: int) -> str:
"""
Format file size in human readable format.
Args:
size_bytes: Size in bytes
Returns:
Formatted size string
"""
if size_bytes == 0:
return "0 B"
size_names = ["B", "KB", "MB", "GB", "TB"]
size = float(size_bytes)
i = 0
while size >= 1024 and i < len(size_names) - 1:
size /= 1024
i += 1
return f"{size:.1f} {size_names[i]}"
def main(clean_first: bool = False) -> None:
"""
Main entry point for gallery generation.
@@ -418,8 +160,14 @@ def main(clean_first: bool = False) -> None:
source_subdirs.append(source.name)
if source_path.is_file() and source_path.suffix == '.pdf':
item = process_plot_files(source_path, source_web_dir)
item = process_plot_files(
CONFIG=CONFIG,
plot_file=source_path,
web_dir=source_web_dir,
)
render_gallery_page(
CONFIG=CONFIG,
template=template,
web_dir=source_web_dir,
items=[item],
subdirs=[],
@@ -436,6 +184,8 @@ def main(clean_first: bool = False) -> None:
print(f"Processed {source.name}: {source.path}")
render_gallery_page(
CONFIG=CONFIG,
template=template,
web_dir=gallery_root,
items=[],
subdirs=source_subdirs,
+230
View File
@@ -0,0 +1,230 @@
import shutil
from typing import Any, Dict
from pathlib import Path
import subprocess
from jinja2 import Template
from utils.metadata import (
resolve_metadata_for_plot,
get_metadata_file_path,
)
from utils.stats import (
calculate_directory_stats,
format_file_size,
)
from utils.config import Config
def process_html_file(
html_file: Path,
web_dir: Path,
current_metadata: Dict[str, Any] = None
) -> dict:
"""
Process HTML plot file, copying it to web directory.
Args:
html_file: Path to the source HTML file
web_dir: Target web directory
current_metadata: Current metadata dictionary for the plot
Returns:
Dictionary containing plot information
"""
web_html = web_dir / html_file.name
if needs_update(html_file, web_html):
shutil.copy2(html_file, web_html)
else:
print(f"Skipping {html_file.name} (up to date)")
# Get source file creation time
source_creation_time = int(html_file.stat().st_ctime)
# Resolve metadata if provided
plot_metadata = {}
if current_metadata is not None:
plot_metadata = resolve_metadata_for_plot(html_file, current_metadata)
return {
"name": html_file.stem,
"html_href": html_file.name,
"is_html": True,
"metadata": plot_metadata,
"creation_time": source_creation_time
}
def process_plot_files(
CONFIG: Config,
plot_file: Path,
web_dir: Path,
current_metadata: Dict[str, Any] = None,
) -> dict:
"""
Process plot files (PDF/PNG or HTML), handling conversion and copying.
Args:
plot_file: Path to the source plot file (PDF or HTML)
web_dir: Target web directory
current_metadata: Current metadata dictionary for the plot
Returns:
Dictionary containing plot information
"""
if plot_file.suffix.lower() == '.html':
return process_html_file(plot_file, web_dir, current_metadata)
# Handle PDF files
png_file = plot_file.with_suffix(".png")
web_pdf = web_dir / plot_file.name
web_png = web_dir / png_file.name
if needs_update(plot_file, web_pdf):
shutil.copy2(plot_file, web_pdf)
else:
print(f"Skipping {plot_file.name} (up to date)")
if not png_file.exists():
convert_pdf_to_png(plot_file, CONFIG=CONFIG)
if needs_update(png_file, web_png):
shutil.copy2(png_file, web_png)
else:
print(f"Skipping {png_file.name} (up to date)")
source_creation_time = int(plot_file.stat().st_ctime)
plot_metadata = {}
if current_metadata is not None:
plot_metadata = resolve_metadata_for_plot(plot_file, current_metadata)
return {
"name": plot_file.stem,
"pdf_href": plot_file.name,
"png_href": png_file.name,
"is_html": False,
"metadata": plot_metadata,
"creation_time": source_creation_time
}
def render_gallery_page(
CONFIG: Config,
template: Template,
web_dir: Path,
items: list,
subdirs: list,
relative_path: Path,
title: str = None,
metadata: dict = None
) -> None:
"""
Unified template rendering for all gallery pages.
Args:
web_dir: Target web directory
items: List of plot items
subdirs: List of subdirectory names
relative_path: Relative path from gallery root
title: Page title (optional)
metadata: Metadata dictionary (optional)
"""
if title is None:
title = "Gallery" if relative_path == Path(
".") else f"Gallery: {relative_path}"
if metadata is None:
metadata = {}
# Calculate statistics
current_stats = calculate_directory_stats(web_dir)
stats = {
"file_count": len(items),
"folder_count": len(subdirs),
"total_size": format_file_size(current_stats["total_size"]),
"total_size_bytes": current_stats["total_size"]
}
# Calculate relative path to assets
if relative_path == Path("."):
assets_path = "../assets"
else:
depth = len(relative_path.parts)
assets_path = "../" * (depth + 1) + "assets"
# For root level, show only directory structure
if relative_path == Path("."):
items = []
output_html = web_dir / "index.html"
with output_html.open("w") as f:
rendered_html = template.render(
title=title,
items=items,
subdirs=subdirs,
relpath=str(relative_path),
paths=CONFIG.paths,
ui=CONFIG.ui,
stats=stats,
folder_metadata=metadata,
assets_path=assets_path,
source_dir=str(web_dir),
metadata_file_path=get_metadata_file_path(web_dir)
)
f.write(rendered_html)
print(f"Generated {output_html}")
def convert_pdf_to_png(pdf_path: Path, CONFIG: Config) -> None:
"""
Convert a PDF file to PNG format using ImageMagick.
Only converts if the PNG doesn't exist or if the PDF is newer than
the PNG (with a 30-second buffer to handle filesystem timing issues).
Args:
pdf_path: Path to the source PDF file
Raises:
subprocess.CalledProcessError: If ImageMagick conversion fails
"""
png_path = pdf_path.with_suffix(".png")
if png_path.exists():
pdf_mtime = pdf_path.stat().st_mtime
png_mtime = png_path.stat().st_mtime
if png_mtime >= (pdf_mtime + 30):
return
else:
print(f"PDF {pdf_path.name} is newer than PNG, reconverting...")
print(f"Converting {pdf_path} to PNG")
subprocess.run([
"convert",
"-density", str(CONFIG.png_dpi),
str(pdf_path),
"-quality", "95",
str(png_path)
], check=True)
def needs_update(source_file: Path, target_file: Path) -> bool:
"""
Check if target file needs updating based on source modification time.
Args:
source_file: Path to the source file
target_file: Path to the target file
Returns:
True if target needs update, False otherwise
"""
if not target_file.exists():
return True
source_mtime = source_file.stat().st_mtime
target_mtime = target_file.stat().st_mtime
return source_mtime > (target_mtime + 30)
+62
View File
@@ -0,0 +1,62 @@
from pathlib import Path
def calculate_directory_stats(directory: Path) -> dict:
"""
Calculate statistics for a directory.
Args:
directory: Path to the directory to analyze
Returns:
Dictionary containing file count, folder count, and total size
"""
stats = {
"file_count": 0,
"folder_count": 0,
"total_size": 0,
"pdf_size": 0,
"png_size": 0,
}
if not directory.exists():
return stats
for item in directory.rglob("*"):
if item.is_file():
stats["file_count"] += 1
size = item.stat().st_size
stats["total_size"] += size
if item.suffix.lower() == '.pdf':
stats["pdf_size"] += size
elif item.suffix.lower() == '.png':
stats["png_size"] += size
elif item.is_dir():
stats["folder_count"] += 1
return stats
def format_file_size(size_bytes: int) -> str:
"""
Format file size in human readable format.
Args:
size_bytes: Size in bytes
Returns:
Formatted size string
"""
if size_bytes == 0:
return "0 B"
size_names = ["B", "KB", "MB", "GB", "TB"]
size = float(size_bytes)
i = 0
while size >= 1024 and i < len(size_names) - 1:
size /= 1024
i += 1
return f"{size:.1f} {size_names[i]}"