feat(server): use poppler for pdf image generation

This commit is contained in:
perf3ct 2025-06-26 20:39:42 +00:00
parent d7aca60733
commit d278d50f3a
No known key found for this signature in database
GPG Key ID: 569C4EEC436F5232
4 changed files with 68 additions and 132 deletions

77
Cargo.lock generated
View File

@ -1143,26 +1143,6 @@ dependencies = [
"crossbeam-utils", "crossbeam-utils",
] ]
[[package]]
name = "console_error_panic_hook"
version = "0.1.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a06aeb73f470f66dcdbf7223caeebb85984942f22f1adb2a088cf9668146bbbc"
dependencies = [
"cfg-if",
"wasm-bindgen",
]
[[package]]
name = "console_log"
version = "1.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "be8aed40e4edbf4d3b4431ab260b63fdc40f5780a4766824329ea0f1eefe3c0f"
dependencies = [
"log",
"web-sys",
]
[[package]] [[package]]
name = "const-oid" name = "const-oid"
version = "0.9.6" version = "0.9.6"
@ -2784,12 +2764,6 @@ dependencies = [
"rawpointer", "rawpointer",
] ]
[[package]]
name = "maybe-owned"
version = "0.3.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4facc753ae494aeb6e3c22f839b158aebd4f9270f55cd3c79906c45476c47ab4"
[[package]] [[package]]
name = "maybe-rayon" name = "maybe-rayon"
version = "0.1.1" version = "0.1.1"
@ -3296,32 +3270,6 @@ dependencies = [
"unicode-normalization", "unicode-normalization",
] ]
[[package]]
name = "pdfium-render"
version = "0.8.33"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dca4101838fe3cb35e5f9dd8b55569ea483e7a342a6853878a5e56fcfdb04cc5"
dependencies = [
"bitflags 2.9.1",
"bytemuck",
"bytes",
"chrono",
"console_error_panic_hook",
"console_log",
"image",
"itertools",
"js-sys",
"libloading",
"log",
"maybe-owned",
"once_cell",
"utf16string",
"vecmath",
"wasm-bindgen",
"wasm-bindgen-futures",
"web-sys",
]
[[package]] [[package]]
name = "peeking_take_while" name = "peeking_take_while"
version = "0.1.2" version = "0.1.2"
@ -3365,12 +3313,6 @@ version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8b870d8c151b6f2fb93e84a13146138f05d02ed11c7e7c54f8826aaaf7c9f184" checksum = "8b870d8c151b6f2fb93e84a13146138f05d02ed11c7e7c54f8826aaaf7c9f184"
[[package]]
name = "piston-float"
version = "1.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ad78bf43dcf80e8f950c92b84f938a0fc7590b7f6866fbcbeca781609c115590"
[[package]] [[package]]
name = "pkcs1" name = "pkcs1"
version = "0.7.5" version = "0.7.5"
@ -3719,7 +3661,6 @@ dependencies = [
"mime_guess", "mime_guess",
"notify", "notify",
"pdf-extract", "pdf-extract",
"pdfium-render",
"quick-xml", "quick-xml",
"raw-cpuid", "raw-cpuid",
"regex", "regex",
@ -5378,15 +5319,6 @@ version = "2.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "daf8dba3b7eb870caf1ddeed7bc9d2a049f3cfdfae7cb521b087cc33ae4c49da" checksum = "daf8dba3b7eb870caf1ddeed7bc9d2a049f3cfdfae7cb521b087cc33ae4c49da"
[[package]]
name = "utf16string"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0b62a1e85e12d5d712bf47a85f426b73d303e2d00a90de5f3004df3596e9d216"
dependencies = [
"byteorder",
]
[[package]] [[package]]
name = "utf8_iter" name = "utf8_iter"
version = "1.0.4" version = "1.0.4"
@ -5477,15 +5409,6 @@ version = "0.2.15"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "accd4ea62f7bb7a82fe23066fb0957d48ef677f6eeb8215f372f52e48bb32426" checksum = "accd4ea62f7bb7a82fe23066fb0957d48ef677f6eeb8215f372f52e48bb32426"
[[package]]
name = "vecmath"
version = "1.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "956ae1e0d85bca567dee1dcf87fb1ca2e792792f66f87dced8381f99cd91156a"
dependencies = [
"piston-float",
]
[[package]] [[package]]
name = "version-compare" name = "version-compare"
version = "0.2.0" version = "0.2.0"

View File

@ -35,7 +35,6 @@ notify = "8"
mime_guess = "2" mime_guess = "2"
tesseract = { version = "0.15", optional = true } tesseract = { version = "0.15", optional = true }
pdf-extract = { version = "0.9", optional = true } pdf-extract = { version = "0.9", optional = true }
pdfium-render = { version = "0.8", optional = true }
image = { version = "0.25", features = ["png", "jpeg", "tiff", "bmp"], optional = true } image = { version = "0.25", features = ["png", "jpeg", "tiff", "bmp"], optional = true }
imageproc = { version = "0.25", optional = true } imageproc = { version = "0.25", optional = true }
thiserror = "2.0" thiserror = "2.0"
@ -60,7 +59,7 @@ testcontainers-modules = { version = "0.12", features = ["postgres"], optional =
[features] [features]
default = ["ocr", "s3"] default = ["ocr", "s3"]
ocr = ["tesseract", "pdf-extract", "pdfium-render", "image", "imageproc", "raw-cpuid"] ocr = ["tesseract", "pdf-extract", "image", "imageproc", "raw-cpuid"]
s3 = ["aws-config", "aws-sdk-s3", "aws-credential-types", "aws-types"] s3 = ["aws-config", "aws-sdk-s3", "aws-credential-types", "aws-types"]
test-utils = ["testcontainers", "testcontainers-modules"] test-utils = ["testcontainers", "testcontainers-modules"]

View File

@ -10,7 +10,7 @@ RUN npm run build
# --- Backend build stage --- # --- Backend build stage ---
FROM rust:1.87-bookworm as backend-builder FROM rust:1.87-bookworm as backend-builder
# Install system dependencies for OCR # Install system dependencies for OCR and PDF processing
RUN apt-get update && apt-get install -y \ RUN apt-get update && apt-get install -y \
tesseract-ocr \ tesseract-ocr \
tesseract-ocr-eng \ tesseract-ocr-eng \
@ -19,6 +19,7 @@ RUN apt-get update && apt-get install -y \
pkg-config \ pkg-config \
libclang-dev \ libclang-dev \
clang \ clang \
poppler-utils \
&& rm -rf /var/lib/apt/lists/* && rm -rf /var/lib/apt/lists/*
WORKDIR /app WORKDIR /app
@ -35,6 +36,7 @@ RUN apt-get update && apt-get install -y \
tesseract-ocr \ tesseract-ocr \
tesseract-ocr-eng \ tesseract-ocr-eng \
ca-certificates \ ca-certificates \
poppler-utils \
&& rm -rf /var/lib/apt/lists/* && rm -rf /var/lib/apt/lists/*
WORKDIR /app WORKDIR /app

View File

@ -320,61 +320,73 @@ impl FileService {
#[cfg(feature = "ocr")] #[cfg(feature = "ocr")]
async fn generate_pdf_thumbnail(&self, file_data: &[u8]) -> Result<Vec<u8>> { async fn generate_pdf_thumbnail(&self, file_data: &[u8]) -> Result<Vec<u8>> {
use pdfium_render::prelude::*; use std::process::Command;
use tokio::fs;
use uuid::Uuid;
// Try to render first page as image using pdfium-render with panic protection // Create a temporary file for the PDF
match catch_unwind(AssertUnwindSafe(|| { let temp_id = Uuid::new_v4();
// Initialize pdfium let temp_pdf_path = format!("/tmp/pdf_thumb_{}.pdf", temp_id);
let pdfium = Pdfium::new( let temp_png_path = format!("/tmp/pdf_thumb_{}.png", temp_id);
Pdfium::bind_to_library(Pdfium::pdfium_platform_library_name_at_path("./"))
.or_else(|_| Pdfium::bind_to_system_library()) // Write PDF data to temporary file
.unwrap_or_else(|_| Pdfium::bind_to_library(Pdfium::pdfium_platform_library_name()).unwrap()) if let Err(e) = fs::write(&temp_pdf_path, file_data).await {
); error!("Failed to write temporary PDF file: {}", e);
return self.generate_placeholder_thumbnail("PDF").await;
// Load PDF document from memory }
let document = pdfium.load_pdf_from_byte_vec(file_data.to_vec(), None)?;
// Use pdftoppm to convert first page to PNG
// Get first page let output = Command::new("pdftoppm")
let page = document.pages().first()?; .arg("-f").arg("1") // First page only
.arg("-l").arg("1") // Last page (same as first)
// Render page to bitmap (200x200 thumbnail size) .arg("-scale-to").arg("200") // Scale to 200px width
let bitmap = page.render_with_config( .arg("-png") // Output as PNG
&PdfRenderConfig::new() .arg(&temp_pdf_path)
.set_target_width(200) .arg(&format!("/tmp/pdf_thumb_{}", temp_id)) // Output prefix
.set_maximum_height(200) .output();
)?;
// Clean up temporary PDF file
// Convert to image format let _ = fs::remove_file(&temp_pdf_path).await;
let width = bitmap.width() as u32;
let height = bitmap.height() as u32; match output {
let buffer = bitmap.as_raw_bytes(); Ok(result) if result.status.success() => {
// pdftoppm adds "-1" to the filename for the first page
// Create RGB image from BGRA buffer let actual_png_path = format!("/tmp/pdf_thumb_{}-1.png", temp_id);
let mut rgb_buffer = Vec::with_capacity((width * height * 3) as usize);
for chunk in buffer.chunks(4) { // Read the generated PNG file
if chunk.len() >= 4 { match fs::read(&actual_png_path).await {
// Convert BGRA to RGB Ok(png_data) => {
rgb_buffer.push(chunk[2]); // R // Clean up temporary PNG file
rgb_buffer.push(chunk[1]); // G let _ = fs::remove_file(&actual_png_path).await;
rgb_buffer.push(chunk[0]); // B
// Convert PNG to JPEG thumbnail
match image::load_from_memory(&png_data) {
Ok(img) => {
// Resize to 200x200 maintaining aspect ratio
let thumbnail = img.resize(200, 200, image::imageops::FilterType::Lanczos3);
// Convert to JPEG
let mut buffer = Vec::new();
let mut cursor = std::io::Cursor::new(&mut buffer);
if thumbnail.write_to(&mut cursor, ImageFormat::Jpeg).is_ok() {
Ok(buffer)
} else {
self.generate_placeholder_thumbnail("PDF").await
}
}
Err(_) => self.generate_placeholder_thumbnail("PDF").await,
}
}
Err(_) => {
let _ = fs::remove_file(&actual_png_path).await;
self.generate_placeholder_thumbnail("PDF").await
}
} }
} }
_ => {
let img = image::ImageBuffer::from_raw(width, height, rgb_buffer) // Clean up any potential PNG files
.ok_or_else(|| anyhow::anyhow!("Failed to create image from buffer"))?; let _ = fs::remove_file(&temp_png_path).await;
let _ = fs::remove_file(&format!("/tmp/pdf_thumb_{}-1.png", temp_id)).await;
let dynamic_img = image::DynamicImage::ImageRgb8(img);
// Convert to JPEG
let mut buffer = Vec::new();
let mut cursor = std::io::Cursor::new(&mut buffer);
dynamic_img.write_to(&mut cursor, image::ImageFormat::Jpeg)?;
Ok(buffer) as anyhow::Result<Vec<u8>>
})) {
Ok(Ok(thumbnail)) => Ok(thumbnail),
Ok(Err(_)) | Err(_) => {
// Fall back to placeholder if PDF rendering fails or panics
self.generate_placeholder_thumbnail("PDF").await self.generate_placeholder_thumbnail("PDF").await
} }
} }