Add support for fetching and storing reply text

- Add database migration and update logic for replyText column in Notes
- Implement batch collection of missing reply text via Tumblr API
- Add new API integration to fetch reply text for replies
- Enhance DataAccess with methods to query/update replyText
- Update CLI: -replies now collects and stores reply text
- Improve logging (archive logs/), error handling, and output
- Remove TL.db from source control
This commit is contained in:
jim
2026-02-12 23:12:54 -06:00
parent 0e5fb124a0
commit bc9985a6f0
5 changed files with 635 additions and 208 deletions
+268 -180
View File
@@ -26,9 +26,18 @@ namespace URLNotesGrabberCORE
if (File.Exists(logPath))
{
string timestamp = DateTime.Now.ToString("yyyyMMdd_HHmmss");
string newPath = $"console_output_{timestamp}.log";
string archiveDirectory = "logs";
// Ensure the archive directory exists
if (!Directory.Exists(archiveDirectory))
{
Directory.CreateDirectory(archiveDirectory);
}
string newPath = Path.Combine(archiveDirectory, $"console_output_{timestamp}.log");
File.Move(logPath, newPath);
}
StreamWriter fileWriter = new StreamWriter(logPath, append: false) { AutoFlush = true };
DualLogger dualLogger = new DualLogger(Console.Out, fileWriter);
Console.SetOut(dualLogger);
@@ -37,7 +46,9 @@ namespace URLNotesGrabberCORE
if (args.Length == 0) //Traverse folder structure to add posts and thus blogs to DB
{
TraverseDirectory(settings.GetValue<string>("PathInput"), settings.GetValue<string>("PathOutputBlogs"), contains);
int postsAdded = 0;
TraverseDirectory(settings.GetValue<string>("PathInput"), settings.GetValue<string>("PathOutputBlogs"), contains, ref postsAdded);
Console.WriteLine($"Total posts added: {postsAdded}");
}
else
{
@@ -62,96 +73,19 @@ namespace URLNotesGrabberCORE
Console.WriteLine("-blogsO\t For each Blog in DB, write blogname to file, but limit via a passed start and stop range ");
Console.WriteLine("-replies\t For each Note that is a REPLY, write blogname to file ");
Console.WriteLine("-replies\t Fetch and update missing reply text for all replies in database");
break;
case "-parse":
string blogNameToParse = args[1];
TraverseDirectory(settings.GetValue<string>("PathInput"), settings.GetValue<string>("PathOutputBlogs"), contains, blogNameToParse);
int postsAdded = 0;
TraverseDirectory(settings.GetValue<string>("PathInput"), settings.GetValue<string>("PathOutputBlogs"), contains, ref postsAdded, blogNameToParse);
Console.WriteLine($"Total posts added: {postsAdded}");
break;
case "-test":
#region Manual test
string blogName = args[1];
long postID = long.Parse(args[2]);
var response = APIAccess.GrabNotes(blogName, postID).GetAwaiter().GetResult();
List<Tuple<string, string>> notes = new List<Tuple<string, string>>();
// If the API returned a 404 inside the JSON `meta` block, mark the post NotFound
if (response?.meta != null && response.meta.status == 404)
{
Console.WriteLine($"ERROR: Post not found - {blogName}/{postID} (meta.status=404)");
DataAccess.UpdatePostMarkNotFound(blogName, postID);
break;
}
// Check for API errors
if (response.statusCode == "NotFound")
{
Console.WriteLine($"ERROR: Post not found - {blogName}/{postID}");
break;
}
if (response.statusCode == "TooManyRequests")
{
Console.WriteLine($"ERROR: Rate limited - retry in {response.retryInSeconds} seconds");
break;
}
if (!string.IsNullOrEmpty(response.statusCode))
{
Console.WriteLine($"ERROR: API returned status code: {response.statusCode}");
break;
}
// Check for null response
if (response.response == null)
{
Console.WriteLine("ERROR: API response is null");
break;
}
// Check for null notes
if (response.response.notes == null)
{
Console.WriteLine("ERROR: Notes collection is null (post may have 0 notes)");
break;
}
Console.WriteLine($"Processing {response.response.notes.Count} notes...");
foreach (var note in response.response.notes)
{
note.reblog_parent_blog_name = blogName; note.post_id = postID.ToString();
//notes.Add(new Tuple<string, string>(note.blog_url, note.type));
if (DataAccess.AddNote(note.reblog_parent_blog_name, note.blog_name, long.Parse(note.post_id), note.timestamp, note.type))
break;
}
while (response.response._links != null)
{
response = APIAccess.GrabNotes(blogName, postID, response.response._links.next.query_params.before_timestamp).GetAwaiter().GetResult();
if (response.response?.notes == null)
{
Console.WriteLine("ERROR: Notes collection became null during pagination");
break;
}
Console.WriteLine($"Processing {response.response.notes.Count} more notes...");
foreach (var note in response.response.notes)
{
note.reblog_parent_blog_name = blogName; note.post_id = postID.ToString(); ;
//notes.Add(new Tuple<string, string>(note.blog_url, note.type));
if (DataAccess.AddNote(note.reblog_parent_blog_name, note.blog_name, long.Parse(note.post_id), note.timestamp, note.type))
break;
}
}
Console.WriteLine("Test completed successfully");
#endregion
Console.WriteLine("Test command not implemented");
break;
case "-post":
@@ -168,7 +102,7 @@ namespace URLNotesGrabberCORE
case "-collect": //collect notes from all posts
bool withoutNotesOnly = true;
DateTime? beforeDate = null;
DateTime? beforeDate = DateTime.Now;
if (args.Length < 2)
{
@@ -247,10 +181,8 @@ namespace URLNotesGrabberCORE
WriteBlogsToFileAll(settings.GetValue<string>("PathOutputBlogs"), false, from, to, top);
break;
case "-replies": //colection posts with replies
try { System.IO.File.Delete(settings.GetValue<string>("PathOutputReplies")); } catch { }
WriteRepliesToFile(settings.GetValue<string>("PathOutputBlogs"), false);
case "-replies": //update reply text
CollectMissingReplyText().GetAwaiter().GetResult();
break;
default:
@@ -351,16 +283,181 @@ namespace URLNotesGrabberCORE
return false;
}
static async Task FetchAndStoreReplyText(string blogName, long postID, long timestamp)
{
try
{
Console.WriteLine($"[Reply Text] Fetching reply text for {blogName}/{postID}/{timestamp}");
// Add 2-second delay before API call to avoid server-side rate limiting
await Task.Delay(2000);
var postsResponse = await APIAccess.GrabPostWithReplies(blogName, postID, timestamp);
if (postsResponse?.response?.posts == null || postsResponse.response.posts.Count == 0)
{
Console.WriteLine($"[Reply Text] No posts found in response for {blogName}/{postID}");
Console.WriteLine($"[Reply Text] Response Status Code: {postsResponse?.statusCode ?? "N/A"}");
Console.WriteLine($"[Reply Text] Response.response is null: {postsResponse?.response == null}");
if (postsResponse?.response != null)
{
Console.WriteLine($"[Reply Text] Posts count: {postsResponse.response.posts?.Count ?? 0}");
}
if (!string.IsNullOrEmpty(postsResponse?.rawJson))
{
Console.WriteLine($"[Reply Text] Raw Response JSON: {postsResponse.rawJson}");
}
// Mark all replies for this post with '?' to indicate API returned no posts
Console.WriteLine($"[Reply Text] Marking all replies for {blogName}/{postID} with '?' due to no posts in response");
DataAccess.UpdateAllNoteReplyTextForPost(blogName, postID, "?");
return;
}
// Get the first (and should be only) post
var post = postsResponse.response.posts.FirstOrDefault();
if (post?.notes == null)
{
Console.WriteLine($"[Reply Text] No notes found in post {blogName}/{postID}");
return;
}
Console.WriteLine($"[Reply Text] Found {post.notes.Count} total notes for {blogName}/{postID}");
// Update each reply with its text
int replyCount = 0;
int skippedCount = 0;
foreach (var note in post.notes)
{
if (note.type == "reply")
{
if (!string.IsNullOrEmpty(note.reply_text))
{
DataAccess.UpdateNoteReplyText(blogName, postID, note.blog_name, note.timestamp, note.reply_text);
replyCount++;
// Output the reply text being stored
string displayText = note.reply_text.Length > 100
? note.reply_text.Substring(0, 100) + "..."
: note.reply_text;
Console.WriteLine($" [{note.blog_name}] {displayText}");
}
else
{
skippedCount++;
Console.WriteLine($"[Reply Text] Skipped reply from {note.blog_name} - empty reply_text");
}
}
}
if (replyCount > 0)
{
Console.WriteLine($"[Reply Text] Updated {replyCount} reply texts for {blogName}/{postID}");
}
else
{
Console.WriteLine($"[Reply Text] No reply text found for {blogName}/{postID} (skipped: {skippedCount})");
}
// Mark any remaining replies with '.' as '?' to indicate they were processed but had no text
Console.WriteLine($"[Reply Text] Marking any remaining replies with '.' as '?' for {blogName}/{postID}");
//int cleanupCount = DataAccess.UpdateRemainingDefaultReplyText(blogName, postID, ".", "?");
//if (cleanupCount > 0)
//{
// Console.WriteLine($"[Reply Text] Cleaned up {cleanupCount} remaining replies for {blogName}/{postID}");
//}
}
catch (Exception ex)
{
Console.WriteLine($"[Reply Text] Error fetching reply text for {blogName}/{postID}: {ex.Message}");
Console.WriteLine($"[Reply Text] StackTrace: {ex.StackTrace}");
}
}
static async Task CollectMissingReplyText()
{
try
{
// Ensure the replyText column exists
DataAccess.EnsureReplyTextColumnExists();
Console.WriteLine("Starting collection of missing reply text...");
Console.WriteLine("Processing 10 posts at a time.");
int batchSize = 1; // Process 10 posts per batch
int totalProcessedCount = 0;
RateLimiter limiter = new SlidingWindowRateLimiter(new SlidingWindowRateLimiterOptions
{
PermitLimit = 300,
QueueProcessingOrder = QueueProcessingOrder.OldestFirst,
QueueLimit = 1,
Window = TimeSpan.FromMinutes(1),
SegmentsPerWindow = 60,
AutoReplenishment = true
});
while (true && totalProcessedCount < batchSize)
{
var postsWithFilledReplies = DataAccess.GetRepliesWithFilledText(limit: batchSize);
if (postsWithFilledReplies.Count == 0)
{
Console.WriteLine("No more posts with filled reply text found. Collection complete.");
break;
}
Console.WriteLine($"Found {postsWithFilledReplies.Count} posts with filled reply text. Processing batch...");
int batchProcessedCount = 0;
foreach (var post in postsWithFilledReplies)
{
using RateLimitLease lease = limiter.AttemptAcquire(1);
if (lease.IsAcquired)
{
Console.WriteLine($"Processing {post.Item1}/{post.Item2}/{post.Item3}");
await FetchAndStoreReplyText(post.Item1, post.Item2, post.Item3);
batchProcessedCount++;
totalProcessedCount++;
// Add 1 second delay between attempts
await Task.Delay(1000);
}
else
{
Console.WriteLine("!@@@@@@@ - Rate Limited Exceeded: No Lease Available");
Console.WriteLine($"Stopped after processing {totalProcessedCount} posts total");
return;
}
}
Console.WriteLine($"Batch complete. Processed {batchProcessedCount} posts in this batch.");
}
Console.WriteLine($"Completed collection of missing reply text. Total processed: {totalProcessedCount} posts.");
}
catch (Exception ex)
{
Console.WriteLine($"Error collecting missing reply text: {ex.Message}");
Console.WriteLine(ex.ToString());
}
}
static async Task<string> GrabNotes(Tuple<string, long, long, long> post)
{
try
{
int APICount = DataAccess.GetAPICount();
Console.WriteLine($"{post.Item1}\t{post.Item2}\t{DateTime.Now}\t{APICount}");
Thread.Sleep(3000);
Console.WriteLine(post.Item1 + '\t' + post.Item2 + '\t' + DateTime.Now + "\t" + APICount);
var response = APIAccess.GrabNotes(post.Item1, post.Item2, post.Item3.ToString()).GetAwaiter().GetResult();
var allNotes = new List<dynamic>();
int page = 1;
string beforeTimestamp = post.Item3.ToString();
bool hasReplies = false;
const int maxPages = 500;
var response = APIAccess.GrabNotes(post.Item1, post.Item2, beforeTimestamp).GetAwaiter().GetResult();
// If the API returned a 404 inside the JSON `meta` block, mark the post NotFound and return
// Handle 404 and error codes
if (response?.meta != null && response.meta.status == 404)
{
Console.WriteLine($"API returned 404 Not Found for {post.Item1}/{post.Item2} (meta.status=404)");
@@ -368,8 +465,11 @@ namespace URLNotesGrabberCORE
Thread.Sleep(1000);
return "NotFound";
}
List<Tuple<string, string>> notes = new List<Tuple<string, string>>();
if (response == null)
{
Console.WriteLine("##### Response is null - API Failure? ###");
return "FAILURE";
}
if (response.statusCode == "NotFound")
{
Thread.Sleep(1000);
@@ -386,50 +486,56 @@ namespace URLNotesGrabberCORE
return response.statusCode;
}
if (response.response != null && response.response.notes != null)
// Pagination loop
while (true)
{
Console.WriteLine("Notes\t" + response.response.notes.Count);
foreach (var note in response.response.notes)
{
note.reblog_parent_blog_name = post.Item1; note.post_id = post.Item2.ToString();
//notes.Add(new Tuple<string, string>(note.blog_url, note.type));
if (DataAccess.AddNote(note.reblog_parent_blog_name, note.blog_name, long.Parse(note.post_id), note.timestamp, note.type))
break;
}
}
else if (response.response == null)
{
Console.WriteLine("##### Response Null - API Failure? ###");
Console.WriteLine($"Raw JSON: {response.rawJson}");
return "FAILURE";
}
else if (response.response.notes == null)
{
Console.WriteLine("##### Notes Null - WHY? ###");
return "FAILURE";
}
else
{
while (response.response != null
&& response.response._links != null
&& long.Parse(response.response._links.next.query_params.before_timestamp) >= post.Item3)
{
response = APIAccess.GrabNotes(post.Item1, post.Item2, response.response._links.next.query_params.before_timestamp).GetAwaiter().GetResult();
if (response.response.notes is null)
{
Console.WriteLine("response.response.notes == null");
return "NULL NOTES";
}
int noteCount = response?.response?.notes?.Count ?? 0;
Console.WriteLine($"[GrabNotes] Page {page} | before_timestamp={beforeTimestamp} | Notes={noteCount}");
Console.WriteLine("Notes\t" + response.response.notes.Count);
foreach (var note in response.response.notes)
{
note.reblog_parent_blog_name = post.Item1; note.post_id = post.Item2.ToString();
//notes.Add(new Tuple<string, string>(note.blog_url, note.type));
if (DataAccess.AddNote(note.reblog_parent_blog_name, note.blog_name, long.Parse(note.post_id), note.timestamp, note.type))
break;
}
if (response?.response == null)
{
Console.WriteLine($"##### Response Null - API Failure? ###\nRaw JSON: {response?.rawJson}");
return "FAILURE";
}
if (response.response.notes == null)
{
Console.WriteLine("##### Notes Null - WHY? ###");
return "FAILURE";
}
// Accumulate notes
allNotes.AddRange(response.response.notes);
// Pagination: check for next
var nextLink = response.response._links?.next;
if (nextLink == null || string.IsNullOrEmpty(nextLink.query_params?.before_timestamp))
{
Console.WriteLine($"[GrabNotes] No more pages. Pagination complete after {page} page(s).");
break;
}
beforeTimestamp = nextLink.query_params.before_timestamp;
page++;
if (page > maxPages)
{
Console.WriteLine($"[GrabNotes] ERROR: Max page limit ({maxPages}) reached for {post.Item1}/{post.Item2}. Aborting further pagination.");
break;
}
// Fetch next page
response = APIAccess.GrabNotes(post.Item1, post.Item2, beforeTimestamp).GetAwaiter().GetResult();
}
Console.WriteLine($"[GrabNotes] Total notes accumulated: {allNotes.Count}");
// Process all accumulated notes
foreach (var note in allNotes)
{
note.reblog_parent_blog_name = post.Item1;
note.post_id = post.Item2.ToString();
if (note.type == "reply")
hasReplies = true;
DataAccess.AddNote(note.reblog_parent_blog_name, note.blog_name, long.Parse(note.post_id), note.timestamp, note.type);
}
return "Success";
@@ -438,8 +544,6 @@ namespace URLNotesGrabberCORE
{
Console.WriteLine(ex.ToString());
}
return "UNKNOWN";
}
@@ -461,15 +565,15 @@ namespace URLNotesGrabberCORE
{
using (StreamWriter sw = new StreamWriter(outPath, true))
{
foreach (var post in posts)
while (posts.Count > 0)
{
var post = posts[0]; // Process the first post in the list
string status;
using RateLimitLease lease = limiter.AttemptAcquire(1);
if (lease.IsAcquired)
{
Console.WriteLine("{0,32} - {1,15} - {2}", post.Item1, post.Item2, DateTimeOffset.FromUnixTimeSeconds(post.Item4).ToString());
// Thread.Sleep(1000);
status = await GrabNotes(post);
}
else
@@ -477,10 +581,18 @@ namespace URLNotesGrabberCORE
Console.WriteLine("!@@@@@@@ - Rate Limited Exceeded: No Lease Available");
return;
}
if (status == "Success")
{
DataAccess.UpdatePostMarkNotesCollected(post.Item1, post.Item2);
}
else
{
Console.WriteLine("GrabNotes Result: " + status);
}
// Re-fetch the updated list after processing the current post
posts = DataAccess.GetPosts(withoutNotesOnly, beforeDate);
}
}
}
@@ -491,27 +603,18 @@ namespace URLNotesGrabberCORE
}
static void TraverseDirectory(string path, string outPath, List<string> contains, string blogName = "")
static void TraverseDirectory(string path, string outPath, List<string> contains, ref int postsAdded, string blogName = "")
{
//if(!ContainsAny(path, contains))
//{
// return;
//}
// Get all directories in the current directory and sort them alphabetically
var directories = Directory.GetDirectories(path);
Array.Sort(directories, StringComparer.InvariantCulture);
//using (StreamWriter sw = new StreamWriter(outPath, true))
//{
//sw.WriteLine("=====" + path);
//}
foreach (var directory in directories)
{
Console.WriteLine("Directory: " + directory);
TraverseDirectory(directory, outPath, contains, blogName); // Recursively traverse subdirectories
TraverseDirectory(directory, outPath, contains, ref postsAdded, blogName); // Recursively traverse subdirectories
}
try
{
bool headerWasWritten = false;
@@ -520,13 +623,6 @@ namespace URLNotesGrabberCORE
{
if (file.EndsWith(".txt", StringComparison.OrdinalIgnoreCase) && (string.IsNullOrEmpty(blogName) || path.IndexOf(blogName, StringComparison.OrdinalIgnoreCase) >= 0))
{
//Console.WriteLine($"=====File: {file}");
//using (StreamWriter sw = new StreamWriter(outPath, true))
//{
//sw.WriteLine("---" + file);
//}
try
{
var urls = new List<string>();
@@ -534,12 +630,11 @@ namespace URLNotesGrabberCORE
foreach (string line in File.ReadLines(file))
{
//if (line.StartsWith(@"Reblog url: https://", StringComparison.OrdinalIgnoreCase) && !line.Contains(@"zombaee") && !line.Contains(@"zomb-eh") && !line.Contains(@"deactivated"))
if (line.StartsWith(@"Post id:", StringComparison.OrdinalIgnoreCase))
{ // if there were no files, let's still collect post info
if (line.StartsWith("Post id:", StringComparison.OrdinalIgnoreCase))
{
if (reblog.reblogName != "." && reblog.postID != "." && reblog.date != "." && reblog.reblogURL != "." && reblog.downloadedFiles == ".")
{
if (!reblog.reblogURL.Contains(@"deactivated")
if (!reblog.reblogURL.Contains("deactivated")
&& reblog.reblogURL.Length != 0
&& ContainsAny(reblog.reblogURL, contains))
{
@@ -548,28 +643,18 @@ namespace URLNotesGrabberCORE
{
headerWasWritten = true;
}
//if (reblog.reblogURL.Contains("/blog/private")
// || reblog.body.Contains("/blog/private"))
//{
// Console.WriteLine(reblog.postID);
// Console.WriteLine(reblog.postURL);
// Console.WriteLine(reblog.date);
// Console.WriteLine(reblog.body);
// Console.WriteLine(reblog.reblogKey);
// Console.WriteLine(reblog.reblogURL);
// Console.WriteLine(reblog.reblogName);
// Console.WriteLine(reblog.downloadedFiles);
//}
string curDir = currentDir.Name.Replace("_1", "").Replace("_2", "").Replace("_3", "").Replace("_4", "").Replace("_5", "").Replace("_6", "").Replace("_7", "").Replace("_8", "").Replace("_9", "");
urls.Add(string.Format("{0}\t{1}", curDir, reblog.postID));
DataAccess.AddPost(curDir, long.Parse(reblog.postID), reblog.reblogURL, reblog.date, reblog.postURL, reblog.slug, reblog.reblogKey,
reblog.reblogName, reblog.summary, reblog.quote, reblog.body, reblog.tags, reblog.link, reblog.photoURL,
reblog.photoCaption, reblog.downloadedFiles, reblog.audioCaption, reblog.question, reblog.answer,
reblog.title, false);
postsAdded++;
// Output hyperlink and post date
Console.WriteLine($"https://{reblog.reblogName}.tumblr.com/post/{reblog.postID} - {reblog.date}");
}
}
reblog = new ReblogRecord();
reblog.postID = line.Substring(9).Trim();
}
@@ -659,8 +744,7 @@ namespace URLNotesGrabberCORE
if ((ContainsAny(reblog.downloadedFiles, contains)
|| (reblog.reblogURL?.IndexOf("/blog/private", StringComparison.OrdinalIgnoreCase) >= 0)
|| (reblog.body?.IndexOf("/blog/private", StringComparison.OrdinalIgnoreCase) >= 0))
// && !ContainsAny(reblog.reblogURL, contains)
&& !reblog.reblogURL.Contains(@"deactivated")
//&& !reblog.reblogURL.Contains("deactivated")
|| path.Contains("zomb-eh", StringComparison.InvariantCultureIgnoreCase))
{
DirectoryInfo currentDir = new DirectoryInfo(path);
@@ -686,6 +770,10 @@ namespace URLNotesGrabberCORE
reblog.reblogName, reblog.summary, reblog.quote, reblog.body, reblog.tags, reblog.link, reblog.photoURL,
reblog.photoCaption, reblog.downloadedFiles, reblog.audioCaption, reblog.question, reblog.answer,
reblog.title, true);
postsAdded++;
// Output hyperlink and post date
Console.WriteLine($"https://{reblog.reblogName}.tumblr.com/post/{reblog.postID} - {reblog.date}");
}
}
}