Merge pull request #6 from CoreFiling/feature/configurable-cli-options

Allow the options provided to pdf2htmlEX to be configured at deployment time

Author: nicolas-duvauchel

.github/workflows/docker.yml

@@ -22,6 +22,12 @@ jobs:
         uses: actions/checkout@v3
         with:
           lfs: true
+      - name: Set up dotnet
+        uses: actions/setup-dotnet@v3
+        with:
+          dotnet-version: "8.x"
+      - name: Unit tests
+        run: dotnet test tests/Unit.Tests/Unit.Tests.csproj
       - name: Set up QEMU
         uses: docker/setup-qemu-action@v2
       - name: Set up Docker Buildx
@@ -32,14 +38,10 @@ jobs:
           context: ./src/Pdf2Html
           load: true
           tags: ${{ env.TEST_TAG }}
-      - name: Set up dotnet
-        uses: actions/setup-dotnet@v3
-        with:
-          dotnet-version: "8.x"
       - name: E2E tests
         run: |
           docker run --rm --detach -p 8080:8080 --name pdf2html ${{ env.TEST_TAG }}
-          dotnet test --filter "FullyQualifiedName=E2E.Tests"
+          dotnet test tests/E2E.Tests/E2E.Tests.csproj
           docker stop pdf2html
       - if: github.ref_name == 'main' || github.ref_type == 'tag'
         name: Login to Docker Hub

CHANGELOG.md

@@ -1,9 +1,10 @@
 # Changelog
 
-## develop
+## 0.2.0
 
 * Update to .net 8.
   * Switch base images to Ubuntu Noble (24.04 LTS).
+* Add optional overrides for command-line arguments passed to `pdf2htmlEX`.
 * Patch and build `pdf2htmlEX` as part of this build process to use `libopenjp` instead of `libjpeg` for JPEG-2000 support.
   * All patches are in this source tree, and are applied to directly to the source of the upstream tag during build.
 * Patch issue with non-breaking spaces in `pdf2HTMLEX`.

README.md

@@ -2,6 +2,28 @@
 
 This project is a lightweight HTTP(S) interface to the [pdf2htmlex library](https://pdf2htmlex.github.io/pdf2htmlEX/).
 
+## Running via Docker
+
+```bash
+docker run -p 8080 corefiling/pdf2html:$version
+```
+
+### Overriding `pdf2htmlEX` options
+
+The command line arguments passed into `pdf2htmlEX` can be overridden by passing in environment variables prefixed by `ConversionOptions__`, e.g:
+
+```bash
+docker run -p 8080 -e ConversionOptions__BgFormat=png -e ConversionOptions__OptimizeText=true corefiling/pdf2html$version
+```
+
+The names of these setting keys are converted to lower-kebab-case arguments, and the values are converted to strings as needed - in the above example, the arguments are converted to `--bg-format=png --optimize-text=0`.
+
+The full list of arguments can be found by running `pdf2htmlEX`:
+
+```bash
+docker run corefiling/pdf2html pdf2htmlEX:$version --help
+```
+
 ## Licensing
 
 Since pdf2htmlex is licensed under the GPL, this project is too (see the LICENSE.TXT file).

pdf2html.sln

@@ -11,6 +11,8 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "tests", "tests", "{C361585C
 EndProject
 Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "E2E.Tests", "tests\E2E.Tests\E2E.Tests.csproj", "{9CAB9112-6B91-4615-A34A-B3C66FD3FAE1}"
 EndProject
+Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "Unit.Tests", "tests\Unit.Tests\Unit.Tests.csproj", "{1174DAD4-FE63-4CF6-9F23-3B9FB6BA409A}"
+EndProject
 Global
 	GlobalSection(SolutionConfigurationPlatforms) = preSolution
 		Debug|Any CPU = Debug|Any CPU
@@ -28,9 +30,14 @@ Global
 		{9CAB9112-6B91-4615-A34A-B3C66FD3FAE1}.Debug|Any CPU.Build.0 = Debug|Any CPU
 		{9CAB9112-6B91-4615-A34A-B3C66FD3FAE1}.Release|Any CPU.ActiveCfg = Release|Any CPU
 		{9CAB9112-6B91-4615-A34A-B3C66FD3FAE1}.Release|Any CPU.Build.0 = Release|Any CPU
+		{1174DAD4-FE63-4CF6-9F23-3B9FB6BA409A}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{1174DAD4-FE63-4CF6-9F23-3B9FB6BA409A}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{1174DAD4-FE63-4CF6-9F23-3B9FB6BA409A}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{1174DAD4-FE63-4CF6-9F23-3B9FB6BA409A}.Release|Any CPU.Build.0 = Release|Any CPU
 	EndGlobalSection
 	GlobalSection(NestedProjects) = preSolution
 		{D3B9B4F8-F097-4F12-AB86-72CAE0B4577C} = {ABE1E425-AA84-46A5-98EA-9B6D622EF8A5}
 		{9CAB9112-6B91-4615-A34A-B3C66FD3FAE1} = {C361585C-8D3B-4CA0-A0BF-DB74DDB00EBE}
+		{1174DAD4-FE63-4CF6-9F23-3B9FB6BA409A} = {C361585C-8D3B-4CA0-A0BF-DB74DDB00EBE}
 	EndGlobalSection
 EndGlobal

pdf2html.sln.DotSettings.user

@@ -0,0 +1,5 @@
+<Solution>
+    <Project Path="src/Pdf2Html/Pdf2Html.csproj" Type="Classic C#" />
+    <Project Path="tests/E2E.Tests/E2E.Tests.csproj" Type="Classic C#" />
+    <Project Path="tests/Unit.Tests/Unit.Tests.csproj" Type="Classic C#" />
+</Solution>

pdf2html.slnx

@@ -0,0 +1,3 @@
+using System.Runtime.CompilerServices;
+
+[assembly: InternalsVisibleTo("Unit.Tests")]

src/Pdf2Html/AssemblyInfo.cs

@@ -1,21 +1,17 @@
 using System.Diagnostics;
 using System.Net.Mime;
 using System.Reflection;
+
 using Microsoft.AspNetCore.Mvc;
 
+using Pdf2Html.Settings;
+
 namespace Pdf2Html.Controllers;
 
 [ApiController]
 [Route("/")]
-public class RootController : ControllerBase
+public class RootController(ILogger<RootController> logger, ConversionOptions conversionOptions) : ControllerBase
 {
-    private readonly ILogger<RootController> _logger;
-
-    public RootController(ILogger<RootController> logger)
-    {
-        _logger = logger;
-    }
-
     [HttpGet]
     public ActionResult Get()
     {
@@ -38,19 +34,19 @@ public async Task<ActionResult> Post()
             await using (var tempFileStream = System.IO.File.Open(inputFile, FileMode.Truncate))
             {
                 await Request.Body.CopyToAsync(tempFileStream);
-                _logger.LogInformation($"Copied {FormatToMb(new FileInfo(inputFile).Length)} to {inputFile}");
+                logger.LogInformation($"Copied {FormatToMb(new FileInfo(inputFile).Length)} to {inputFile}");
             }
 
-            _logger.LogInformation("Starting conversion...");
+            logger.LogInformation("Starting conversion...");
             var (success, logs) = await ConvertAsync(inputFile, outputFile);
 
             if (!success)
             {
-                _logger.LogError("Conversion failed");
+                logger.LogError("Conversion failed");
                 return StatusCode(StatusCodes.Status500InternalServerError, new { pdf2htmlEX = new { logs } });
             }
 
-            _logger.LogInformation($"Conversion completed ({FormatToMb(new FileInfo(outputFile).Length)})");
+            logger.LogInformation($"Conversion completed ({FormatToMb(new FileInfo(outputFile).Length)})");
             return File(await System.IO.File.ReadAllBytesAsync(outputFile), MediaTypeNames.Text.Html);
         }
         finally
@@ -63,11 +59,10 @@ public async Task<ActionResult> Post()
     private async Task<(bool Success, ICollection<string> logs)> ConvertAsync(string inputFile, string outputFile)
     {
         using var p = new Process();
-        const string conversionOptions = "--embed-javascript=0 --process-outline=0 --printing=0 --bg-format=svg --svg-node-count-limit=100 --decompose-ligature 1 --tounicode 1";
         p.StartInfo = new ProcessStartInfo
         {
             FileName = "pdf2htmlEX",
-            Arguments = $"{conversionOptions} --dest-dir={Path.GetDirectoryName(outputFile)} {inputFile} {Path.GetFileName(outputFile)}",
+            Arguments = $"{conversionOptions.CommandLineArguments} --dest-dir={Path.GetDirectoryName(outputFile)} {inputFile} {Path.GetFileName(outputFile)}",
             CreateNoWindow = true,
             RedirectStandardOutput = true,
             RedirectStandardError = true
@@ -83,7 +78,7 @@ void AddLog(string? log)
             }
 
             logs.Add(log);
-            _logger.LogInformation(log);
+            logger.LogInformation(log);
         }
 
         p.OutputDataReceived += (_, e) => AddLog(e.Data);
@@ -97,8 +92,5 @@ void AddLog(string? log)
         return (p.ExitCode == 0, logs);
     }
 
-    private static string FormatToMb(long bytesLength)
-    {
-        return (bytesLength / 1024.0 / 1024.0).ToString("0.00 MB");
-    }
+    private static string FormatToMb(long bytesLength) => (bytesLength / 1024.0 / 1024.0).ToString("0.00 MB");
 }

src/Pdf2Html/Controllers/RootController.cs

@@ -1,10 +1,20 @@
+using Pdf2Html.Settings;
+
+using System.Diagnostics;
+using System.Reflection;
+
 var builder = WebApplication.CreateBuilder(args);
 builder.Logging.ClearProviders();
 builder.Logging.AddConsole();
 
 // Add services to the container.
 builder.Services.AddControllers();
+builder.Services.AddSingleton<ConversionOptions>();
 
 var app = builder.Build();
+var versionInfo = FileVersionInfo.GetVersionInfo(Assembly.GetExecutingAssembly().Location);
+app.Logger.LogInformation($"Starting {versionInfo.ProductName} {versionInfo.ProductVersion}");
+app.Logger.LogInformation($"Using pdf2htmlEX command line arguments: {app.Services.GetService<ConversionOptions>()!.CommandLineArguments}");
+
 app.MapControllers();
 app.Run();

src/Pdf2Html/Program.cs

@@ -0,0 +1,16 @@
+using System.Text.RegularExpressions;
+
+namespace Pdf2Html.Settings;
+
+public class ConversionOptions(IConfiguration configuration)
+{
+    public string CommandLineArguments { get; } = ToCommandLineArguments(configuration.GetSection("ConversionOptions").AsEnumerable());
+
+    internal static string ToCommandLineArguments(IEnumerable<KeyValuePair<string, string?>> options) =>
+        string.Join(' ', options.Where(kvp => kvp.Value != null).Select(kvp => $"--{ToKebabCase(kvp.Key.Replace("ConversionOptions:", ""))}={ValueToString(kvp.Value!)}"));
+
+    private static string ValueToString(string value) => bool.TryParse(value, out var boolValue) ? (boolValue ? "1" : "0") : value;
+
+    private static string ToKebabCase(string value) =>
+        Regex.Replace(value, "(?<!^)([A-Z][a-z]|(?<=[a-z])[A-Z0-9])", "-$1", RegexOptions.Compiled).Trim().ToLower();
+}