jim/URLNotesGrabberCore
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: add single-blog filter to -ingest

Browse Source 
`-ingest <blogname>` now restricts the run to one blog's folder, mirroring
the existing -parse <blogname> ergonomics. `-ingest` with no arg still
processes every blog under appSettings:PathTTRoot (or PathInput fallback).

Breaking vs 3aff849: the first positional arg is interpreted as a blog
name, not a path. Configure the root via appSettings:PathTTRoot.

Co-Authored-By: Claude Opus 4.7 <[email protected]>
This commit is contained in:
jim
2026-05-18 15:36:42 -05:00
co-authored by Claude Opus 4.7
parent 3aff849216
commit cf3f97ddc4
2 changed files with 19 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
URLNotesGrabberCORE/IngestMode.cs
+18 -7
View File
@@ -10,13 +10,16 @@ namespace URLNotesGrabberCORE
{
public static int Run(IConfiguration config, string[] args)
{
string? rootPath = args.Length > 0 ? args[0] : config["appSettings:PathTTRoot"];
string? targetBlog = args.Length > 0 ? args[0]?.Trim() : null;
if (string.IsNullOrWhiteSpace(targetBlog)) targetBlog = null;
string? rootPath = config["appSettings:PathTTRoot"];
if (string.IsNullOrWhiteSpace(rootPath))
rootPath = config["appSettings:PathInput"];
if (string.IsNullOrWhiteSpace(rootPath))
{
Console.WriteLine("Ingest: no root path provided. Set appSettings:PathTTRoot, appSettings:PathInput, or pass a path after -ingest.");
Console.WriteLine("Ingest: no root path configured. Set appSettings:PathTTRoot or appSettings:PathInput.");
return 1;
}
@@ -43,6 +46,7 @@ namespace URLNotesGrabberCORE
Console.WriteLine($"========== Ingest Settings ==========");
Console.WriteLine($"Root path: {rootPath}");
Console.WriteLine($"Blog filter: {(targetBlog == null ? "(all blogs)" : targetBlog)}");
Console.WriteLine($"=====================================");
var txtFiles = new List<string>();
@@ -66,6 +70,7 @@ namespace URLNotesGrabberCORE
Console.WriteLine($"Processing {txtFiles.Count} .txt file(s)...");
int filesProcessed = 0;
int filesSkipped = 0;
int postsTouched = 0;
try
@@ -77,14 +82,20 @@ namespace URLNotesGrabberCORE
{
try
{
filesProcessed++;
if (filesProcessed % 50 == 0 || filesProcessed == 1)
Console.WriteLine($"[{filesProcessed}/{txtFiles.Count}] {Path.GetFileName(file)}");
string rawBlogName = Path.GetFileName(Path.GetDirectoryName(file) ?? "unknown");
string blogName = Regex.Replace(rawBlogName, @"_\d+$", "");
string postType = Path.GetFileNameWithoutExtension(file);
if (targetBlog != null && !string.Equals(blogName, targetBlog, StringComparison.OrdinalIgnoreCase))
{
filesSkipped++;
continue;
}
filesProcessed++;
if (filesProcessed % 50 == 0 || filesProcessed == 1)
Console.WriteLine($"[{filesProcessed}/{txtFiles.Count}] {Path.GetFileName(file)}");
string currentPostId = "";
var currentPostData = new Dictionary<string, string>(StringComparer.OrdinalIgnoreCase);
@@ -158,7 +169,7 @@ namespace URLNotesGrabberCORE
DataAccess.RestoreImportModePragmas();
}
Console.WriteLine($"\nIngest complete. Files processed: {filesProcessed}. Posts touched: {postsTouched}.");
Console.WriteLine($"\nIngest complete. Files processed: {filesProcessed}. Files skipped (blog filter): {filesSkipped}. Posts touched: {postsTouched}.");
return 0;
}
URLNotesGrabberCORE/Program.cs
+1 -1
View File
@@ -181,7 +181,7 @@ namespace URLNotesGrabberCORE
Console.WriteLine("-api [section]\t Use a specific API settings section from appsettings.json (e.g. TumblrApi3)");
Console.WriteLine("-ingest [path]\t Ingest Tumblr .txt exports under path (or appSettings:PathTTRoot) into TL.db");
Console.WriteLine("-ingest [blogname]\t Ingest Tumblr .txt exports from appSettings:PathTTRoot into TL.db (all blogs, or single blog if name given)");
Console.WriteLine("-output\t Export posts from TL.db back to .txt files in each blog's TTFolderPath");