> For the complete documentation index, see [llms.txt](https://help.brieftech.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.brieftech.ai/reference/embedded-file-extraction.md). # Embedded file extraction BriefTech can detect embedded content inside uploaded files and make it available for extraction. This page lists every supported container format, explains how detection works, and describes the behaviour of the extraction process. ## Supported container formats BriefTech reports embedded content at **depth 1** for any format in the tables below — the immediate children of the container, not recursively nested content. Inner containers that are themselves extracted appear as new documents and can be extracted in turn. ### Email and mailbox containers The highest-value containers in litigation collections. BriefTech extracts each attachment or message as a separate document. | Format | Extension(s) | MIME type | Embedded content detected | | ------------------- | ---------------- | -------------------------------- | -------------------------------------------------- | | Outlook PST mailbox | `.pst` | `application/vnd.ms-outlook-pst` | All messages and attachments in the mailbox | | Outlook OST mailbox | `.ost` | `application/vnd.ms-outlook-pst` | All messages and attachments — see note below | | Outlook email | `.msg` | `application/vnd.ms-outlook` | Email attachments | | Internet email | `.eml` | `message/rfc822` | Email attachments | | Unix/Mac mailbox | `.mbox` | `application/mbox` | Each message, extracted as individual `.eml` files | | MHTML web archive | `.mht`, `.mhtml` | `message/rfc822` | Embedded resources (images, stylesheets, frames) | > **OST caveat:** OST files created by Outlook 2013 and later use a newer internal format that is only partially supported. If your OST produces fewer extracted items than expected, export to `.pst` or individual `.msg` files before uploading. ### Archive containers | Format | Extension(s) | MIME type | Embedded content detected | | --------------- | ------------------------------------------------ | -------------------------------- | -------------------------- | | ZIP archive | `.zip` | `application/zip` | All files in the archive | | 7-Zip archive | `.7z` | `application/x-7z-compressed` | All files in the archive | | RAR archive | `.rar` | `application/x-rar-compressed` | All files in the archive | | TAR archive | `.tar`, `.tar.gz`, `.tgz`, `.tar.bz2`, `.tar.xz` | `application/x-tar` and variants | All files in the archive | | Gzip stream | `.gz` | `application/gzip` | The single compressed file | | BZip2 stream | `.bz2` | `application/x-bzip2` | The single compressed file | | XZ stream | `.xz` | `application/x-xz` | The single compressed file | | LZMA stream | `.lzma` | `application/x-lzma` | The single compressed file | | Unix compress | `.Z` | `application/x-compress` | The single compressed file | | ARJ archive | `.arj` | `application/x-arj` | All files in the archive | | Unix AR archive | `.ar` | `application/x-archive` | All files in the archive | | CPIO archive | `.cpio` | `application/x-cpio` | All files in the archive | | Java archive | `.jar`, `.war`, `.ear` | `application/java-archive` | All files in the archive | ### Documents with embedded content These formats only show a non-zero embedded-file count when the document itself contains embedded file attachments or OLE objects. A standard PDF with no attached files, or a Word document with no embedded objects, will not appear in the Extractable view. | Format | Extension(s) | MIME type | Embedded content detected | | ------------------------- | ------------ | --------------------------------------------------------------------------- | --------------------------------------------------- | | PDF | `.pdf` | `application/pdf` | Embedded file attachments and PDF portfolio members | | Word (OOXML) | `.docx` | `application/vnd.openxmlformats-officedocument.wordprocessingml.document` | Embedded OLE objects and attached files | | Excel (OOXML) | `.xlsx` | `application/vnd.openxmlformats-officedocument.spreadsheetml.sheet` | Embedded OLE objects and attached files | | PowerPoint (OOXML) | `.pptx` | `application/vnd.openxmlformats-officedocument.presentationml.presentation` | Embedded OLE objects and attached files | | Word (legacy) | `.doc` | `application/msword` | Embedded OLE objects | | Excel (legacy) | `.xls` | `application/vnd.ms-excel` | Embedded OLE objects | | PowerPoint (legacy) | `.ppt` | `application/vnd.ms-powerpoint` | Embedded OLE objects | | OpenDocument text | `.odt` | `application/vnd.oasis.opendocument.text` | Embedded objects | | OpenDocument spreadsheet | `.ods` | `application/vnd.oasis.opendocument.spreadsheet` | Embedded objects | | OpenDocument presentation | `.odp` | `application/vnd.oasis.opendocument.presentation` | Embedded objects | | EPUB e-book | `.epub` | `application/epub+zip` | Component documents and media | ### Partial support These formats are recognised and processed, but have known limitations. | Format | Extension(s) | Limitation | | ------------------- | ---------------- | -------------------------------------------------------------------------------------------------------------------------------------- | | Outlook OST (2013+) | `.ost` | Newer OST format is only partially parsed; some messages may not be detected. Export to PST or MSG before uploading for full coverage. | | SQLite database | `.sqlite`, `.db` | Not available in BriefTech's current configuration. | | Android package | `.apk` | ZIP layer is extracted (component files visible), but Android-specific structures within are not further parsed. | | iOS app package | `.ipa` | Same as `.apk` — ZIP layer only. | | Debian package | `.deb` | Outer container is extracted; inner archives are not further unpacked in a single pass. | ## Detection After upload, BriefTech's file processing pipeline inspects every uploaded file to read its metadata. For supported container formats, BriefTech reports the count and names of embedded files at depth 1. This information is stored with the document and reflected in the Extractable view of the Files tab tab. Detection runs automatically — no action is required. A file that contains embedded content shows a non-zero value in the 📎 column (hover the column header for its "Number of embedded files" tooltip). {% hint style="warning" %} **Size limit on detection:** BriefTech reads a file's embedded listing only while the file is within its processing size limit: * **PDFs** — up to **1 GB** * **Most other types** (7z, RAR, TAR, GZ, PST, MBOX, EML/MSG) — up to **500 MB** * **ZIP and modern Office files** (.docx, .xlsx, .pptx) — **no size limit** A container larger than its limit is rejected before its contents are read, so it never shows an embedded-file count or appears in the Extractable view. Split or extract a very large archive or mailbox locally before uploading. {% endhint %} Archive and mailbox containers (ZIP, TAR, 7z, PST, MBOX, and similar) are handled differently from ordinary documents: * They cannot be rendered to a PDF or OCR'd. * Detection still records their embedded file listing — the count and names of members at depth 1 — which is what makes them appear in the Extractable view. * The container file itself does not finish processing: with no PDF rendition to produce, it halts at the conversion stage, is flagged as an unsupported type, and appears in the Exceptions view. * The content of individual members becomes searchable only after you run extraction on the container. ## The Extractable view The Extractable view in the Files tab tab shows files that: * have at least one detected embedded file, * have not yet been extracted, **and** * are not archived or in the bin. Once extraction has run on a file, it no longer appears in this view. The view itself is hidden from the left panel when there are no extractable files, and when shown its entry carries a badge with the number of files currently waiting to be extracted. ## Extraction Extraction is a one-time, on-demand action. To trigger it, select one or more eligible files and click Extract ZIP & Attachments in the toolbar. See [Extract files from a ZIP or email attachment](/how-to-guides/process/extract-zip-attachments.md) for step-by-step instructions. ### Two ways to extract: everything, or attachments only Two extract actions are offered, in the toolbar and the right-click menu: * Extract ZIP & Attachments unpacks every **named** embedded file at depth 1 — every file in an archive (folder entries themselves are not turned into documents), or every named part of an email including images embedded in its body. A part that carries no filename is skipped; the one exception is a message inside a mailbox, which is named `message_0001.eml`, `message_0002.eml`, and so on. * Extract Attachments Only unpacks **only genuine attachments** and skips images embedded in an email's body. * Two kinds of body image are skipped: parts the email marks as inline, and images referenced from the message body that a mailer mislabelled as attachments. * For a container with no message body — a ZIP or other archive — the two actions behave identically. * Where neither test matches, nothing is treated as a body image and this action extracts everything the other one would. If the source cannot be parsed as an email at all, no body reference can be matched, so only parts the email itself marks as inline are still skipped. Both actions: * consume the file's single extraction, so only the first one run takes effect; * are visible only while a selected file still has embedded content waiting, and are **hidden — not greyed out** — when nothing is eligible; * render as icon-only buttons whose name shows as a tooltip, on the Files tab and the Review tab alike. The number shown beside each action counts the **selected files that are eligible**, not the attachments inside them. For documents processed before this release, that number is an upper bound: an older email whose only embedded content is inline body images is still counted as eligible, yet *Extract Attachments Only* produces nothing from it. ### Permissions Only users with the **Owner** or **Editor** role on the matter can run extraction. ### Inline vs background | Total size of selected files | Execution mode | | ---------------------------- | --------------------------------------------------------------------------------- | | Up to about 10.5 MB | Runs immediately; a success or warning notification appears when done | | More than about 10.5 MB | Runs as a background job; a progress dialog opens and the job continues if closed | Add up the sizes BriefTech shows in the **Size** column — that is the figure this threshold applies to. ### What happens during extraction For each embedded file: 1. BriefTech writes the file's raw bytes into storage as a new document record. 2. The new document is assigned a unique ID and linked to the source file as its parent. 3. The new document is placed in the source file's folder. If the container held its own folder structure, that structure is recreated as subfolders inside it. 4. The source file is marked as extracted so it cannot be extracted again. 5. The new document enters the standard processing pipeline (detect → convert → OCR → parse → process → linearise → finalise) independently. ### Already-extracted files A file that has already been extracted cannot be extracted again. If such a file is included in a selection, it is silently skipped — the extract action counts and processes only the files that still have embedded content waiting, so an already-extracted file is simply left out. ### Empty archives A container with no embedded files has nothing to extract, so it never appears as extractable. Files with zero detected embedded content don't show up in the Extractable view, and the extract action does not appear at all unless at least one selected file still has embedded content waiting to be unpacked. ## After extraction Extracted documents appear in the matter alongside all other documents. They are full documents: they can be reviewed, indexed, moved to folders, labelled, and included in bundles. Like any other document, an extracted document is charged the first time you commit it, and it is not charged again after that. Extracting, detecting, converting and the automatic OCR that runs during processing consume no credits. See [How credits work](/concepts/how-credits-work.md). The source file remains in the matter after extraction. It retains its own document record and is available for download. ## How extracted families appear in the document list After extraction, a source file and the documents unpacked from it form a **family**: the source is the *parent* and each extracted document is a *child*. In the document list on the Files and Review tabs, a fixed column on the far left shows this relationship: * Parent marks a parent — a source file that other documents were extracted from. Hover over it to see how many were extracted (for example, "3 attached files", or "1 attached file" for a single one). * Child marks a child — a document that was extracted from a parent. Hovering over any row in a family highlights all the other documents in the same family, so you can see at a glance which came from the same source. This column appears only when the current view contains at least one family; if nothing in view has extracted children, it is hidden. ## Detaching a document from its family A family is created automatically — you never group documents by hand — so BriefTech also lets you break the link when a document shouldn't stay tied to its parent. In the Document Index, open the More actions menu on a **child** document (one that was extracted from a parent) and choose Detach from parent. BriefTech asks you to confirm, because this can't be undone: > Detach this document from its parent? It will become a standalone document and cannot be grouped back again. Once you confirm, the document becomes standalone: it leaves the family, its child marker disappears, and the parent's attached-file count drops by one. **Detaching is permanent** — a standalone document cannot be regrouped into its original family. You can detach a child from the Document Index in two ways (both ask you to confirm first): * Open the More actions menu on the child and choose Detach from parent. * In **Edit** mode, drag the child's row out of its parent's group. Dragging also detaches several children at once — select the rows, then drag them out together. {% hint style="success" %} **Select a whole family at once.** On any document that belongs to a family, the More actions menu also offers Select family — and Deselect family once selected — to select every document in that family at once, useful before a bulk action. {% endhint %} ## Related
The document lifecycle: from upload to readyWhat happens to a file between upload and the moment it can be used./pages/u8eIgAY8I81LcXEFoaCo
Extract files from a ZIP or email attachmentUnpack the contents of a ZIP archive or email's attachments into separate documents in your matter./pages/zOfyfAZ3rohV5ecVxlWv
Reprocess a failed documentRe-run a document through the processing pipeline after it failed./pages/kntArxiUSk94F5Le5P36
Background JobsStatus indicators, columns, and row actions in the Background Jobs dialog./pages/REaGxjKXuLlDxbm28MDE
Your First MatterA guided, end-to-end first run — take a practice matter from sample files all the way to a downloaded court bundle./pages/JElpS8iDIgC3KPHk5auM
--- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://help.brieftech.ai/reference/embedded-file-extraction.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.