Skip to content

armgit / Filee

Public Imported from GitHub
0

Free, open-source file converter. Convert 263 formats offline by drag & drop — images, PDF, Word, Excel, PowerPoint, HWP/HWPX, video, audio, e-books and archives. No upload, no Office needed.

Merge main into feature/ocr

cre8ive-marketing9993b9a34836
diff --cc README.md
index 6a390ba6,d22ff3eb..4ad3a991
--- a/README.md
+++ b/README.md
@@@ -169,9 -215,7 +215,7 @@@ It connects to the internet only
- to ask GitHub (`api.github.com`) whether a newer release exists, shortly after start and then every few hours.
Turn this off in Settings → About ("Check for updates automatically").
- - to download the conversion engines you choose to install, from the official locations pinned in
- [`engines.json`](src/Filee.Engines/Infrastructure/engines.json), and the text recognition (OCR) languages a preset
- uses, pinned in [`ocr-languages.json`](src/Filee.Engines/Infrastructure/ocr-languages.json).
-- to download the conversion engines you choose to install, from the official locations pinned in [`engines.json`](src/Filee.Engines/Infrastructure/engines.json).
++- to download the conversion engines you choose to install, from the official locations pinned in [`engines.json`](src/Filee.Engines/Infrastructure/engines.json), and the text recognition (OCR) languages a preset uses, pinned in [`ocr-languages.json`](src/Filee.Engines/Infrastructure/ocr-languages.json).
- when you click a link, which opens your web browser.
## Sponsor
diff --cc src/Filee.Core/Settings/AppSettings.cs
index c9856e03,3eed0429..0e732dd0
--- a/src/Filee.Core/Settings/AppSettings.cs
+++ b/src/Filee.Core/Settings/AppSettings.cs
@@@ -112,7 -112,7 +112,7 @@@ public sealed class DonutSetting
public sealed class AppSettings
{
/// <summary>Current schema version. Increase when the format changes and add a migration.</summary>
-- public const int CurrentSchemaVersion = 39;
++ public const int CurrentSchemaVersion = 40;
public int SchemaVersion { get; set; } = CurrentSchemaVersion;
diff --cc src/Filee.Core/Settings/UserDataStore.cs
index 1bdfae95,6b8aa4d8..048dc125
--- a/src/Filee.Core/Settings/UserDataStore.cs
+++ b/src/Filee.Core/Settings/UserDataStore.cs
@@@ -794,13 -794,9 +794,16 @@@ internal static class SettingsMigration
// v38: Tajima embroidery designs (.dst) learn the vector donut; custom claims stay put.
if (from < 38)
changed |= LearnExtensions(profiles, out _, ["dst"]);
- // v39: "Searchable PDF" (text recognition), after "Compress PDF" as in a new library, on no donut.
- if (from < 39 && !presets.Any(p => p.Id == "searchable-pdf"))
+ // v39: PowerShell scripts (.ps1) learn the text donut; custom claims stay put.
+ if (from < 39)
+ changed |= LearnExtensions(profiles, out _, ["ps1"]);
++ // v40: "Searchable PDF" (text recognition), after "Compress PDF" as in a new library, on no donut.
++ if (from < 40 && !presets.Any(p => p.Id == "searchable-pdf"))
+ {
+ var at = presets.FindIndex(p => p.Id == "compress-pdf");
+ presets.Insert(at < 0 ? presets.Count : at + 1, BuiltInData.OcrPresets().Single());
+ changed = true;
+ }
return changed;
}
diff --cc tests/Filee.Core.Tests/OcrRouteTests.cs
index cb8023e9,00000000..e321eee8
mode 100644,000000..100644
--- a/tests/Filee.Core.Tests/OcrRouteTests.cs
+++ b/tests/Filee.Core.Tests/OcrRouteTests.cs
@@@ -1,116 -1,0 +1,116 @@@
+// "Recognize text" (issue #288): Tesseract's OCR edges start a route only when the preset asks for them, and the
+// route goes on from the searchable PDF.
+
+using Filee.Core.Conversion;
+using Filee.Core.Presets;
+using Filee.Core.Settings;
+
+namespace Filee.Core.Tests;
+
+public class OcrRouteTests
+{
+ private static readonly ConversionEdge[] OcrEdges =
+ [new("png", "pdf", Ocr: true), new("jpg", "pdf", Ocr: true), new("pdf", "pdf", Ocr: true)];
+
+ private static string Steps(Route? route) => string.Join(", ", route!.Steps.Select(s => $"{s.Converter.Id}:{s.From}→{s.To}"));
+
+ private static RoutePlanner Planner() => new([
+ new FakeConverter("tesseract", OcrEdges),
+ new FakeConverter("magick", new ConversionEdge("heic", "png"), new ConversionEdge("png", "pdf"), new ConversionEdge("jpg", "pdf")),
+ new FakeConverter("docx", new ConversionEdge("pdf", "docx")),
+ new FakeConverter("text", new ConversionEdge("pdf", "txt")),
+ ]);
+
+ [Fact]
+ public void Ocr_edges_are_never_part_of_an_ordinary_route()
+ {
+ var planner = Planner();
+
+ Assert.Equal("magick:png→pdf", Steps(planner.Plan("png", "pdf")));
+ Assert.Null(planner.Plan("pdf", "pdf")); // no ordinary PDF → PDF edge here
+ }
+
+ [Fact]
+ public void Asking_for_ocr_starts_with_tesseract_and_goes_on_from_its_pdf()
+ {
+ var planner = Planner();
+
+ Assert.Equal("tesseract:jpg→pdf", Steps(planner.Plan("jpg", "pdf", ocr: true)));
+ Assert.Equal("tesseract:pdf→pdf, docx:pdf→docx", Steps(planner.Plan("pdf", "docx", ocr: true)));
+ Assert.Equal("tesseract:png→pdf, text:pdf→txt", Steps(planner.Plan("png", "txt", ocr: true)));
+ // A picture Tesseract can't open becomes a PNG first.
+ Assert.Equal("magick:heic→png, tesseract:png→pdf, docx:pdf→docx", Steps(planner.Plan("heic", "docx", ocr: true)));
+ }
+
+ [Fact]
+ public void Without_tesseract_there_is_no_ocr_route()
+ {
+ var planner = new RoutePlanner([new FakeConverter("magick", new ConversionEdge("png", "pdf"))]);
+
+ Assert.Null(planner.Plan("png", "pdf", ocr: true));
+ Assert.False(planner.CanConvert("png", "pdf", ocr: true));
+ }
+
+ [Fact]
+ public void Text_is_recognized_in_pdfs_and_pictures_for_document_targets()
+ {
+ var preset = new Preset { TargetFormat = "docx", Pdf = { Ocr = true } };
+
+ Assert.True(preset.RecognizesTextOf("pdf"));
+ Assert.True(preset.RecognizesTextOf("jpg"));
+ Assert.False(preset.RecognizesTextOf("docx")); // already text
+ Assert.False(new Preset { TargetFormat = "png", Pdf = { Ocr = true } }.RecognizesTextOf("pdf"));
+ Assert.False(new Preset { TargetFormat = "docx" }.RecognizesTextOf("pdf"));
+ }
+
+ [Fact]
+ public async Task A_job_with_ocr_runs_tesseract_before_the_document_writer()
+ {
+ using var dir = new TempDir();
+ var scan = dir.File("scan.jpg");
+ var tesseract = new FakeRecorder("tesseract", OcrEdges);
+ var writer = new FakeRecorder("docx", new ConversionEdge("pdf", "docx"));
+ var magick = new FakeRecorder("magick", new ConversionEdge("jpg", "pdf"));
+ await using var queue = new JobQueue(new ConverterCatalog([tesseract, writer, magick]));
+
+ var job = new ConversionJob([scan], new Preset { TargetFormat = "docx", Pdf = { Ocr = true } }, "Test");
+ await queue.RunAsync(job);
+
+ Assert.Equal(JobState.Completed, job.State);
+ Assert.Equal([("jpg", "pdf", false)], tesseract.Steps);
+ Assert.Equal([("pdf", "docx", false)], writer.Steps);
+ Assert.Empty(magick.Steps);
+ }
+
+ [Fact]
+ public async Task Pdfs_merged_into_one_are_recognized_first()
+ {
+ using var dir = new TempDir();
+ var a = dir.File("a.pdf");
+ var b = dir.File("b.jpg");
+ var tesseract = new FakeRecorder("tesseract", OcrEdges);
+ var merger = new FakeMerger();
+ await using var queue = new JobQueue(new ConverterCatalog([tesseract]), merger);
+
+ var job = new ConversionJob([a, b], new Preset { TargetFormat = "pdf", Pdf = { Ocr = true, MergeIntoSingle = true } }, "Test");
+ await queue.RunAsync(job);
+
+ Assert.Equal(JobState.Completed, job.State);
+ Assert.Equal([("jpg", "pdf", false), ("pdf", "pdf", false)], tesseract.Steps.OrderBy(s => s.From));
+ }
+
+ [Fact]
- public void A_version_38_library_gets_searchable_pdf_after_compress_pdf()
++ public void A_version_39_library_gets_searchable_pdf_after_compress_pdf()
+ {
+ var presets = BuiltInData.CreatePresets();
+ presets.RemoveAll(p => p.Id == "searchable-pdf");
+ var profiles = BuiltInData.CreateProfiles();
+
- Assert.True(SettingsMigrations.ApplyToLibrary(38, presets, profiles));
++ Assert.True(SettingsMigrations.ApplyToLibrary(39, presets, profiles));
+
+ var ids = presets.Select(p => p.Id).ToList();
+ Assert.Equal(ids.IndexOf("compress-pdf") + 1, ids.IndexOf("searchable-pdf"));
+ Assert.True(presets.Single(p => p.Id == "searchable-pdf").Pdf.Ocr);
- Assert.False(SettingsMigrations.ApplyToLibrary(39, presets, profiles));
++ Assert.False(SettingsMigrations.ApplyToLibrary(40, presets, profiles));
+ }
+}