feat: merge ThreeTxtFileHelper into URLNotesGrabberCORE

Folds the standalone ThreeTxtFileHelper tool into URLNotesGrabberCORE so text-file ingest/output/correct lives alongside the API scraper. Adds new flags -ingest, -output, -correct (with -apply), -updatepaths, and a one-time -importposts <posts.db> migration. Schema: Blogs.TTFolderPath and Posts.PostType are added by an idempotent migration. On (BlogName, PostID) collisions, content columns are overwritten while engagement columns (ByLikes, RootBlogName, RootURL, HasNotesGathered, NotFound, NotesGatheredDateTime, Likes*) are preserved. Co-Authored-By: Claude Opus 4.7 <[email protected]>
2026-05-18 12:43:17 -05:00
parent 5781e121d2
commit 3aff849216
10 changed files with 1419 additions and 1 deletions
@@ -0,0 +1,70 @@
+using System.Text.Json;
+
+namespace URLNotesGrabberCORE
+{
+    // Port of ThreeTxtFileHelper/UpdateBlogPaths.cs. Reads .tumblr / .tmblrpriv metadata
+    // files from a root\Index folder and populates Blogs.TTFolderPath in TL.db.
+    public static class UpdateBlogPathsRunner
+    {
+        public static int Run(string rootPath)
+        {
+            if (string.IsNullOrWhiteSpace(rootPath))
+            {
+                Console.WriteLine("UpdateBlogPaths: rootPath is required.");
+                return 1;
+            }
+
+            DataAccess.EnsureTTFileHelperColumnsExist();
+
+            string indexPath = Path.Combine(rootPath, "Index");
+            if (!Directory.Exists(indexPath))
+            {
+                Console.WriteLine($"Index folder not found at: {indexPath}");
+                return 1;
+            }
+
+            Console.WriteLine($"Scanning Index folder: {indexPath}");
+
+            var blogFiles = Directory.GetFiles(indexPath, "*.tumblr")
+                .Concat(Directory.GetFiles(indexPath, "*.tmblrpriv"))
+                .ToList();
+
+            Console.WriteLine($"Found {blogFiles.Count} blog metadata files");
+
+            int updatedCount = 0;
+
+            foreach (var blogFile in blogFiles)
+            {
+                try
+                {
+                    string blogName = Path.GetFileNameWithoutExtension(blogFile);
+                    string jsonContent = File.ReadAllText(blogFile);
+                    using JsonDocument doc = JsonDocument.Parse(jsonContent);
+                    JsonElement root = doc.RootElement;
+
+                    if (root.TryGetProperty("FileDownloadLocation", out JsonElement locationElement))
+                    {
+                        string? fileDownloadLocation = locationElement.GetString();
+                        if (!string.IsNullOrWhiteSpace(fileDownloadLocation))
+                        {
+                            DataAccess.SetBlogTTFolderPath(blogName, fileDownloadLocation);
+                            updatedCount++;
+                            Console.WriteLine($"Updated {blogName}: {fileDownloadLocation}");
+                        }
+                    }
+                    else
+                    {
+                        Console.WriteLine($"No FileDownloadLocation found in {blogFile}");
+                    }
+                }
+                catch (Exception ex)
+                {
+                    Console.WriteLine($"Error processing {blogFile}: {ex.Message}");
+                }
+            }
+
+            Console.WriteLine($"\nUpdated {updatedCount} blogs with TTFolderPath");
+            return 0;
+        }
+    }
+}