public FolderWatcherManager(WebLibraryDetail _library)
{
web_library_detail = new TypedWeakReference <WebLibraryDetail>(_library);
// Load any pre-existing watched filenames
bool file_exists;
// Utilities.LockPerfTimer l1_clk = Utilities.LockPerfChecker.Start();
lock (filenames_processed_lock)
{
// l1_clk.LockPerfTimerStop();
file_exists = File.Exists(Filename_Store);
}
if (file_exists)
{
Logging.Info("Loading memory of files that we watched previously.");
// Utilities.LockPerfTimer l2_clk = Utilities.LockPerfChecker.Start();
lock (filenames_processed_lock)
{
// l2_clk.LockPerfTimerStop();
foreach (string filename in File.ReadAllLines(Filename_Store))
{
filenames_processed.Add(filename);
}
}
}
if (ConfigurationManager.IsEnabled(nameof(FolderWatcher)))
{
managed_thread_index = Utilities.Maintainable.MaintainableManager.Instance.RegisterHeldOffTask(TaskDaemonEntryPoint, 30 * 1000, extra_descr: $".Lib({LibraryRef})");
}
}
/// <summary>
/// Removes the entry.
/// </summary>
/// <param name="target">The target object.</param>
public void Remove(DependencyObject target)
{
TypedWeakReference <DependencyObject> singleOrDefault =
_inner.Keys.SingleOrDefault(x => ReferenceEquals(x.Target, target));
if (singleOrDefault != null)
{
if (_inner[singleOrDefault].IsAlive)
{
_inner[singleOrDefault].Target.Dispose();
}
_inner.Remove(singleOrDefault);
}
}
internal PDFDocument(WebLibraryDetail web_library_detail, DictionaryBasedObject dictionary, PDFAnnotationList prefetched_annotations_for_document = null)
{
this.library = new TypedWeakReference <WebLibraryDetail>(web_library_detail);
this.dictionary = dictionary;
// process any prefetched annotations that we may have as usual:
if (prefetched_annotations_for_document != null)
{
annotations = prefetched_annotations_for_document;
lock (access_lock)
{
dirtyNeedsReindexing = (prefetched_annotations_for_document.Count > 0);
}
}
}
private static Image GetImage(string name)
{
Image result = null;
TypedWeakReference <Image> reference;
if (s_imageCache.TryGetValue(name, out reference) && reference.IsAlive)
{
result = reference.Target;
}
if (result == null)
{
result = Image.FromFile(name, true);
s_imageCache[name] = new TypedWeakReference <Image>(result);
}
return(result);
}
protected virtual void Dispose(bool disposing)
{
Logging.Debug("LibraryIndex::Dispose({0}) @{1}", disposing, dispose_count);
WPFDoEvents.SafeExec(() =>
{
if (dispose_count == 0)
{
// Get rid of managed resources
// Utilities.LockPerfTimer l1_clk = Utilities.LockPerfChecker.Start();
lock (word_index_manager_lock)
{
// l1_clk.LockPerfTimerStop();
word_index_manager?.Dispose();
word_index_manager = null;
}
//this.library?.Dispose();
}
});
WPFDoEvents.SafeExec(() =>
{
//this.word_index_manager = null;
library = null;
});
WPFDoEvents.SafeExec(() =>
{
//Utilities.LockPerfTimer l4_clk = Utilities.LockPerfChecker.Start();
lock (pdf_documents_in_library_lock)
{
//l4_clk.LockPerfTimerStop();
pdf_documents_in_library?.Clear();
pdf_documents_in_library = null;
}
});
++dispose_count;
}
public void Dispose()
{
#if DIAG
Logging.Debug("FolderWatcherManager::Dispose() @{0}", ++dispose_count);
#endif
WPFDoEvents.SafeExec(() =>
{
//Utilities.LockPerfTimer l1_clk = Utilities.LockPerfChecker.Start();
lock (folder_watcher_records_lock)
{
//l1_clk.LockPerfTimerStop();
// Dispose of all the folder watchers
foreach (var folder_watcher_record in folder_watcher_records)
{
folder_watcher_record.Value.folder_watcher.Dispose();
}
folder_watcher_records.Clear();
}
});
WPFDoEvents.SafeExec(() =>
{
Utilities.Maintainable.MaintainableManager.Instance.CleanupEntry(managed_thread_index);
});
WPFDoEvents.SafeExec(() =>
{
//Library.Dispose();
web_library_detail = null;
});
WPFDoEvents.SafeExec(() =>
{
// Utilities.LockPerfTimer l2_clk = Utilities.LockPerfChecker.Start();
lock (filenames_processed_lock)
{
// l2_clk.LockPerfTimerStop();
filenames_processed.Clear();
}
});
}
protected virtual void Dispose(bool disposing)
{
Logging.Debug("FolderWatcher::Dispose({0}) @{1}", disposing, dispose_count);
WPFDoEvents.SafeExec(() =>
{
if (dispose_count == 0)
{
// Get rid of managed resources
file_system_watcher.EnableRaisingEvents = false;
file_system_watcher.Dispose();
}
file_system_watcher = null;
});
WPFDoEvents.SafeExec(() =>
{
folder_watcher_manager = null;
//library.TypedTarget.Dispose();
web_library_detail = null;
});
WPFDoEvents.SafeExec(() =>
{
tags.Clear();
});
WPFDoEvents.SafeExec(() =>
{
configured_folder_to_watch = null;
aspiring_folder_to_watch = null;
});
WPFDoEvents.SafeExec(() =>
{
watch_stats.Reset(null);
});
++dispose_count;
}
public FolderWatcher(FolderWatcherManager _folder_watcher_manager, WebLibraryDetail _library, string folder_to_watch, string _tags)
{
folder_watcher_manager = new TypedWeakReference <FolderWatcherManager>(_folder_watcher_manager);
web_library_detail = new TypedWeakReference <WebLibraryDetail>(_library);
aspiring_folder_to_watch = folder_to_watch;
tags = TagTools.ConvertTagBundleToTags(_tags);
configured_folder_to_watch = null;
watch_stats = new WatchStatistics();
file_system_watcher = new FileSystemWatcher();
file_system_watcher.IncludeSubdirectories = true;
file_system_watcher.Filter = "*.pdf";
file_system_watcher.Changed += file_system_watcher_Changed;
file_system_watcher.Created += file_system_watcher_Created;
configured_folder_to_watch = null;
FolderContentsHaveChanged = false;
file_system_watcher.Path = null;
file_system_watcher.EnableRaisingEvents = false;
}
public AITagManager(Library library)
{
this.library = new TypedWeakReference <Library>(library);
current_ai_tags_record = new AITags();
// Attempt to load the existing tags
try
{
if (File.Exists(Filename_Store))
{
Stopwatch clk = Stopwatch.StartNew();
Logging.Info("+Loading AutoTags");
current_ai_tags_record = SerializeFile.ProtoLoad <AITags>(Filename_Store);
Logging.Info("-Loading AutoTags (time spent: {0} ms)", clk.ElapsedMilliseconds);
}
}
catch (Exception ex)
{
Logging.Warn(ex, "There was a problem loading existing AutoTags");
}
}
public AITagManager(WebLibraryDetail web_library_detail)
{
WPFDoEvents.AssertThisCodeIs_NOT_RunningInTheUIThread();
this.web_library_detail = new TypedWeakReference <WebLibraryDetail>(web_library_detail);
current_ai_tags_record = new AITags();
// Attempt to load the existing tags
try
{
if (File.Exists(Filename_Store))
{
Stopwatch clk = Stopwatch.StartNew();
Logging.Info("+Loading AutoTags");
current_ai_tags_record = SerializeFile.ProtoLoad <AITags>(Filename_Store);
Logging.Info("-Loading AutoTags (time spent: {0} ms)", clk.ElapsedMilliseconds);
}
}
catch (Exception ex)
{
Logging.Warn(ex, "There was a problem loading existing AutoTags");
}
}
internal PDFDocument_ThreadUnsafe(Library library, DictionaryBasedObject dictionary)
{
this.library = new TypedWeakReference <Library>(library);
this.dictionary = dictionary;
}
public ExpeditionManager(WebLibraryDetail web_library_detail)
{
this.web_library_detail = new TypedWeakReference <WebLibraryDetail>(web_library_detail);
}
public RecentlyReadManager(Library library)
{
this.library = new TypedWeakReference <Library>(library);
}
public WordList GetOCRText(int page, bool queue_for_ocr)
{
lock (texts)
{
// First check our cache
{
TypedWeakReference <WordList> word_list_weak;
texts.TryGetValue(page, out word_list_weak);
if (null != word_list_weak)
{
WordList word_list = word_list_weak.TypedTarget;
if (null != word_list)
{
return(word_list);
}
}
}
// Then check for an existing SINGLE file
{
string filename = pdf_render_file_layer.MakeFilename_TextSingle(page);
try
{
if (File.Exists(filename))
{
// Get this ONE page
Dictionary <int, WordList> word_lists = WordList.ReadFromFile(filename, page);
WordList word_list = word_lists[page];
texts[page] = new TypedWeakReference <WordList>(word_list);
return(word_list);
}
}
catch (Exception ex)
{
Logging.Warn(ex, "There was an error loading the OCR text for {0} page {1}.", document_fingerprint, page);
FileTools.Delete(filename);
}
}
// Then check for an existing GROUP file
{
string filename = pdf_render_file_layer.MakeFilename_TextGroup(page, TEXT_PAGES_PER_GROUP);
try
{
if (File.Exists(filename))
{
Dictionary <int, WordList> word_lists = WordList.ReadFromFile(filename);
foreach (var pair in word_lists)
{
texts[pair.Key] = new TypedWeakReference <WordList>(pair.Value);
}
TypedWeakReference <WordList> word_list_weak;
texts.TryGetValue(page, out word_list_weak);
if (null != word_list_weak)
{
WordList word_list = word_list_weak.TypedTarget;
if (null != word_list)
{
return(word_list);
}
}
}
}
catch (Exception ex)
{
Logging.Warn(ex, "There was an error loading the OCR text group for {0} page {1}.", document_fingerprint, page);
FileTools.Delete(filename);
}
}
}
// If we get this far then the text was not available so queue extraction
if (queue_for_ocr)
{
// If we have never tried the GROUP version before, queue for it
string filename = pdf_render_file_layer.MakeFilename_TextGroup(page, TEXT_PAGES_PER_GROUP);
if (!File.Exists(filename))
{
PDFTextExtractor.Instance.QueueJobGroup(new PDFTextExtractor.Job(this, page, TEXT_PAGES_PER_GROUP));
}
else
{
PDFTextExtractor.Instance.QueueJobSingle(new PDFTextExtractor.Job(this, page, TEXT_PAGES_PER_GROUP));
}
}
return(null);
}
/// <summary>
/// Returns the OCR words on the page. Null if the words are not yet available.
/// The page will be queued for OCRing if they are not available...
/// Page is 1 based...
/// </summary>
/// <param name="page"></param>
/// <returns></returns>
public WordList GetOCRText(int page, bool queue_for_ocr = true)
{
//Utilities.LockPerfTimer l1_clk = Utilities.LockPerfChecker.Start();
lock (texts_lock)
{
//l1_clk.LockPerfTimerStop();
// First check our cache
{
TypedWeakReference <WordList> word_list_weak;
texts.TryGetValue(page, out word_list_weak);
if (null != word_list_weak)
{
WordList word_list = word_list_weak.TypedTarget;
if (null != word_list)
{
return(word_list);
}
}
}
// Then check for an existing SINGLE file
{
string filename = pdf_render_file_layer.MakeFilename_TextSingle(page);
try
{
if (File.Exists(filename))
{
// Get this ONE page
Dictionary <int, WordList> word_lists = WordList.ReadFromFile(filename, page);
WordList word_list = word_lists[page];
if (null == word_list)
{
throw new Exception(String.Format("No words on page {0} in OCR file {1}", page, filename));
}
texts[page] = new TypedWeakReference <WordList>(word_list);
return(word_list);
}
}
catch (Exception ex)
{
Logging.Warn(ex, "There was an error loading the OCR text for {0} page {1}.", document_fingerprint, page);
FileTools.Delete(filename);
}
}
// Then check for an existing GROUP file
{
string filename = pdf_render_file_layer.MakeFilename_TextGroup(page);
try
{
if (File.Exists(filename))
{
Dictionary <int, WordList> word_lists = WordList.ReadFromFile(filename);
foreach (var pair in word_lists)
{
texts[pair.Key] = new TypedWeakReference <WordList>(pair.Value);
}
TypedWeakReference <WordList> word_list_weak;
texts.TryGetValue(page, out word_list_weak);
if (null != word_list_weak)
{
WordList word_list = word_list_weak.TypedTarget;
if (null != word_list)
{
return(word_list);
}
}
}
}
catch (Exception ex)
{
Logging.Warn(ex, "There was an error loading the OCR text group for {0} page {1}.", document_fingerprint, page);
FileTools.Delete(filename);
}
}
}
// If we get this far then the text was not available so queue extraction
if (queue_for_ocr)
{
// If we have never tried the GROUP version before, queue for it
string filename = pdf_render_file_layer.MakeFilename_TextGroup(page);
PDFTextExtractor.Job job = new PDFTextExtractor.Job(this, page);
if (!File.Exists(filename) && PDFTextExtractor.Instance.JobGroupHasNotFailedBefore(job))
{
PDFTextExtractor.Instance.QueueJobGroup(job);
}
else
{
PDFTextExtractor.Instance.QueueJobSingle(job);
}
}
return(null);
}
public BlackWhiteListManager(WebLibraryDetail web_library_detail)
{
this.web_library_detail = new TypedWeakReference <WebLibraryDetail>(web_library_detail);
}
public RecentlyReadManager(WebLibraryDetail web_library_detail)
{
this.web_library_detail = new TypedWeakReference <WebLibraryDetail>(web_library_detail);
}
/// <summary>
/// Returns the OCR words on the page. Null if the words are not yet available.
/// The page will be queued for OCRing if they are not available...
/// Page is 1 based...
/// </summary>
/// <param name="page"></param>
/// <returns></returns>
public WordList GetOCRText(int page, bool queue_for_ocr = true)
{
WPFDoEvents.AssertThisCodeIs_NOT_RunningInTheUIThread();
if (page > PageCount || page < 1)
{
// dump stacktrace with this one so we know who instigated this out-of-bounds request.
//
// Boundary issue was discovered during customer log file analysis (log files courtesy of Chris Hicks)
try
{
throw new ArgumentException($"INTERNAL ERROR: requesting page text for page {page} which lies outside the detected document page range 1..{PageCount} for PDF document {Fingerprint}");
}
catch (Exception ex)
{
Logging.Error(ex);
}
}
//Utilities.LockPerfTimer l1_clk = Utilities.LockPerfChecker.Start();
lock (texts_lock)
{
//l1_clk.LockPerfTimerStop();
// First check our cache
{
TypedWeakReference <WordList> word_list_weak;
texts.TryGetValue(page, out word_list_weak);
if (null != word_list_weak)
{
WordList word_list = word_list_weak.TypedTarget;
if (null != word_list)
{
return(word_list);
}
}
}
// Then check for an existing SINGLE file
{
string filename = MakeFilename_TextSingle(page);
try
{
if (File.Exists(filename))
{
// Get this ONE page
Dictionary <int, WordList> word_lists = WordList.ReadFromFile(filename, page);
WordList word_list = word_lists[page];
if (null == word_list)
{
throw new Exception(String.Format("No words on page {0} in OCR file {1}", page, filename));
}
texts[page] = new TypedWeakReference <WordList>(word_list);
return(word_list);
}
}
catch (Exception ex)
{
Logging.Warn(ex, "There was an error loading the OCR text for {0} page {1}.", Fingerprint, page);
FileTools.Delete(filename);
}
}
// Then check for an existing GROUP file
{
string filename = MakeFilename_TextGroup(page);
try
{
if (File.Exists(filename))
{
Dictionary <int, WordList> word_lists = WordList.ReadFromFile(filename);
foreach (var pair in word_lists)
{
texts[pair.Key] = new TypedWeakReference <WordList>(pair.Value);
}
TypedWeakReference <WordList> word_list_weak;
texts.TryGetValue(page, out word_list_weak);
if (null != word_list_weak)
{
WordList word_list = word_list_weak.TypedTarget;
if (null != word_list)
{
return(word_list);
}
}
}
}
catch (Exception ex)
{
Logging.Warn(ex, "There was an error loading the OCR text group for {0} page {1}.", Fingerprint, page);
FileTools.Delete(filename);
}
}
}
// If we get this far then the text was not available so queue extraction
if (queue_for_ocr)
{
// If we have never tried the GROUP version before, queue for it
string filename = MakeFilename_TextGroup(page);
PDFTextExtractor.Job job = new PDFTextExtractor.Job(this, page);
if (!File.Exists(filename) && PDFTextExtractor.Instance.JobGroupHasNotFailedBefore(job))
{
PDFTextExtractor.Instance.QueueJobGroup(job);
}
else
{
PDFTextExtractor.Instance.QueueJobSingle(job);
}
}
return(null);
}
public ExpeditionManager(Library library)
{
this.library = new TypedWeakReference <Library>(library);
}
public PasswordManager(WebLibraryDetail web_library_detail)
{
this.web_library_detail = new TypedWeakReference <WebLibraryDetail>(web_library_detail);
}
public LibraryIndex(Library library)
{
this.library = new TypedWeakReference <Library>(library);
// postpone INIT phase...
}
public BlackWhiteListManager(Library library)
{
this.library = new TypedWeakReference <Library>(library);
}
public PasswordManager(Library library)
{
this.library = new TypedWeakReference <Library>(library);
}
public LibraryIndex(WebLibraryDetail web_library_detail)
{
this.library = new TypedWeakReference <Library>(library);
// postpone INIT phase...
}
internal PDFDocument(WebLibraryDetail web_library_detail)
{
this.library = new TypedWeakReference <WebLibraryDetail>(web_library_detail);
dictionary = new DictionaryBasedObject();
}
