Title, Provider, tags from paperless

frontend/src/utils/paperless.ts

@@ -0,0 +1,198 @@
+import { PaperlessDocument } from "../types";
+
+const CORRESPONDENT_KEYS = ["correspondent_name", "correspondent__name", "correspondent"];
+
+const TAG_SOURCES = ["tag_names", "tags__name", "tagNames", "tags", "tagLabels", "tag_details"];
+
+export function extractPaperlessTitle(document: PaperlessDocument): string | null {
+  const record = document as Record<string, unknown>;
+  const value = record.title;
+  if (typeof value === "string" && value.trim().length > 0) {
+    return value.trim();
+  }
+  return null;
+}
+
+export function extractPaperlessProvider(document: PaperlessDocument): string | null {
+  const record = document as Record<string, unknown>;
+
+  for (const key of CORRESPONDENT_KEYS) {
+    const value = record[key];
+    if (typeof value === "string" && value.trim().length > 0) {
+      return value.trim();
+    }
+    if (value && typeof value === "object") {
+      const candidate =
+        (value as Record<string, unknown>).name ??
+        (value as Record<string, unknown>).title ??
+        (value as Record<string, unknown>).label ??
+        (value as Record<string, unknown>).value;
+      if (typeof candidate === "string" && candidate.trim().length > 0) {
+        return candidate.trim();
+      }
+    }
+  }
+
+  const metadata = record.metadata as Record<string, unknown> | undefined;
+  if (metadata) {
+    if (Array.isArray(metadata)) {
+      const providerFromArray = extractProviderFromArray(metadata);
+      if (providerFromArray) {
+        return providerFromArray;
+      }
+    }
+
+    const keys = Object.keys(metadata);
+    for (const key of keys) {
+      if (!key.toLowerCase().includes("correspondent")) {
+        continue;
+      }
+      const value = metadata[key];
+      if (typeof value === "string" && value.trim().length > 0) {
+        return value.trim();
+      }
+      if (value && typeof value === "object") {
+        const candidate =
+          (value as Record<string, unknown>).name ??
+          (value as Record<string, unknown>).title ??
+          (value as Record<string, unknown>).label ??
+          (value as Record<string, unknown>).value;
+        if (typeof candidate === "string" && candidate.trim().length > 0) {
+          return candidate.trim();
+        }
+      }
+    }
+  }
+
+  return null;
+}
+
+export function extractPaperlessTags(document: PaperlessDocument): string[] {
+  const record = document as Record<string, unknown>;
+  const names = new Set<string>();
+
+  for (const key of TAG_SOURCES) {
+    const source = record[key];
+    if (!Array.isArray(source)) {
+      maybeCollectFromValue(source, names);
+      continue;
+    }
+    collectFromArray(source, names);
+  }
+
+  const metadata = record.metadata as Record<string, unknown> | Array<unknown> | undefined;
+  if (metadata) {
+    if (Array.isArray(metadata)) {
+      for (const entry of metadata) {
+        maybeCollectFromValue(entry, names);
+      }
+    } else {
+      for (const key of TAG_SOURCES) {
+        const value = metadata[key];
+        if (Array.isArray(value)) {
+          collectFromArray(value, names);
+        } else {
+          maybeCollectFromValue(value, names);
+        }
+      }
+
+      const metaTags = metadata.tags ?? metadata.tag_list ?? metadata.TagList;
+      if (Array.isArray(metaTags)) {
+        collectFromArray(metaTags, names);
+      } else if (typeof metaTags === "string") {
+        collectFromString(metaTags, names);
+      }
+
+      for (const [key, value] of Object.entries(metadata)) {
+        if (!key.toLowerCase().includes("tag")) {
+          continue;
+        }
+        if (Array.isArray(value)) {
+          collectFromArray(value, names);
+        } else if (typeof value === "string") {
+          collectFromString(value, names);
+        }
+      }
+    }
+  }
+
+  return Array.from(names);
+}
+
+function collectFromArray(items: Array<unknown>, names: Set<string>) {
+  for (const item of items) {
+    if (typeof item === "string" && item.trim().length > 0) {
+      names.add(item.trim());
+      continue;
+    }
+    maybeCollectFromValue(item, names);
+  }
+}
+
+function maybeCollectFromValue(value: unknown, names: Set<string>) {
+  if (!value) {
+    return;
+  }
+  if (typeof value === "string") {
+    collectFromString(value, names);
+    return;
+  }
+  if (Array.isArray(value)) {
+    collectFromArray(value, names);
+    return;
+  }
+  const record = value as Record<string, unknown>;
+  const candidate =
+    record.name ??
+    record.label ??
+    record.title ??
+    record.value ??
+    record.slug;
+  if (typeof candidate === "string" && candidate.trim().length > 0) {
+    names.add(candidate.trim());
+    return;
+  }
+  const objectValues = Object.values(record);
+  for (const nested of objectValues) {
+    if (typeof nested === "string" && nested.trim().length > 0) {
+      names.add(nested.trim());
+    } else if (nested && typeof nested === "object") {
+      maybeCollectFromValue(nested, names);
+    }
+  }
+}
+
+function collectFromString(value: string, names: Set<string>) {
+  const parts = value
+    .split(/[,;]+/)
+    .map((part) => part.trim())
+    .filter(Boolean);
+  for (const part of parts) {
+    names.add(part);
+  }
+}
+
+function extractProviderFromArray(entries: Array<unknown>): string | null {
+  for (const entry of entries) {
+    if (!entry || typeof entry !== "object") {
+      continue;
+    }
+    const record = entry as Record<string, unknown>;
+    const lowerKeys = Object.keys(record).map((key) => key.toLowerCase());
+    const hasCorrespondentKey = lowerKeys.some((key) => key.includes("correspondent"));
+    const slug = (record.slug ?? record.key ?? record.field) as string | undefined;
+    const slugMatches = typeof slug === "string" && slug.toLowerCase().includes("correspondent");
+    if (!hasCorrespondentKey && !slugMatches) {
+      continue;
+    }
+    const candidate =
+      record.value ??
+      record.name ??
+      record.label ??
+      record.title;
+    if (typeof candidate === "string" && candidate.trim().length > 0) {
+      return candidate.trim();
+    }
+  }
+  return null;
+}