OmniDocHow to export or copy PDF directory
Explain clearly that PDF directories come in two forms: text and bookmarks, how to copy and export them to Word or Excel, how to perform OCR recognition on scanned documents first, and give a checklist after exporting.
Key points of this article
- First distinguish whether the table of contents is text printed on the page or a PDF bookmark. The methods of removing the two are different.
- You can select text on the table of contents page, directly copy or export it to Word, and then organize it into a table.
- The scanned table of contents is a picture and needs to be OCR recognized first. After recognition, the title and page number are checked one by one.
- After exporting, check the level indentation, page number correspondence and missing lines, and check a few lines to compare with the original text for confirmation.
What many people want is not to add a new table of contents to the PDF, but to take out the existing table of contents: use it as a reference structure for the paper, organize the chapter list of the report, or put the list of terms of the contract into a table. This type of demand is in the opposite direction to "generating a directory page", and the approach is also different.
This article is written for people who need to extract the table of contents from PDF. After reading, you can determine which type of directory you have, choose the right method of copying or exporting, and check whether there are any serialization, missing lines or misaligned page numbers after getting the results.
The "table of contents" in PDF usually exists in two ways. One is the text printed on a certain page, which is a common table of contents page; the other is the bookmark (outline) in the sidebar of the reader, which records the jump relationship of chapters and is not necessarily displayed on the page. Whether they can be copied and how to export them are very different.
Can the PDF table of contents be copied? One sentence conclusion
Yes, as long as the directory contains selectable text, it can be taken out directly; if it is a picture, OCR recognition must be done first; if there are only sidebar bookmarks, read according to the bookmark. The following table gives methods based on common situations.
| your situation | Recommended practices | reason |
|---|---|---|
| Text can be selected with the mouse on the table of contents page | Directly select and copy, or export to Word for finishing | The text layer exists and the content can be taken out as it is. |
| The catalog page is a picture and cannot be selected. | Do OCR recognition first, then copy or export | The words in the picture do not have a text layer and must be recognized first |
| There is no table of contents on the page, only the chapter tree is seen in the sidebar | Read bookmarks level by level, or read the bookmark structure programmatically | Bookmarks are independent navigation data and are not necessarily printed on the page. |
| Need to check page numbers one by one | Export to Excel or paste into table, put titles and page numbers in columns | The table facilitates line-by-line checking |
| The table of contents is very long and spans several pages. | Select in pages, or export the directory page separately before processing | If you select too many at once, it is easy to get lost in the text. |
Before you start: Confirm these points first
- Confirm where the table of contents is: Turn to the previous pages and see if there is a page marked "Table of Contents". If yes, process it according to the table of contents page; if not, open the sidebar of the reader to view the bookmarks. The contents of the two may not be exactly the same, whichever one you need shall prevail.
- Test whether text can be selected: Use the mouse to drag from the beginning to the end of a line in the directory. If it can be highlighted, it means there is a text layer; if it can only pull out a box or there is no response, it means it is a picture. This step determines which path to take later.
- Confirm that the file can be opened normally: Password-protected PDFs require a password before they can be copied or exported. This is a file permissions issue and should be resolved with file owner authorization and not used to bypass protections set by others.
- Determine the target format: If you want to paste into a document, use Word; if you want to check item by item or make statistics, use Excel; you only need a few lines to take notes and copy them directly. If you choose the right format, there will be much less organizing later.
- Confirm whether the use requires permission: The table of contents is usually just structural information, but if you want to reprint or publish the entire document, you still need to confirm the copyright permission.
Steps to export PDF table of contents in OmniDoc
If the table of contents is selectable text, it can be exported using the PDF editing function of the OmniDoc desktop client. The client can be downloaded at download.omnidoc.top and supports Windows, macOS and Linux.
- Install and open the desktop client: If it has not been installed yet, first go to the download page to obtain the version of the corresponding system, and then open it after installation.
- Import PDF: Add the file to be processed to the editor and wait for the document to be fully loaded.
- Turn to the table of contents page: Make sure that the title and page number of the table of contents are displayed on the page, and do not skip the folded or paged parts.
- OCR the scanned document first: If the table of contents cannot be selected, use the OCR function to identify the page where the table of contents is located or the entire document. After completion, check whether the recognized text is correct.
- Export to Word: Select Export to Word and save the file. In this way, the directory content will become editable text for easy deletion.
- Organize in Word: Delete headers, footers and irrelevant page numbers, and unify the indentation or numbering of each level of headings.
- Export to Excel when a table is required: If you want to compare items one by one, choose to export to Excel, or paste the items into the table, with the title in the first column and the page number in the second column.
When you only need a few lines of content, you don’t have to go through the entire process: select the table of contents line in the reader, copy it, and paste it into a text editor to view it.
Check each item after completion
After you get the results, do not use them directly, check them according to the checklist below.
| Check items | What do you think? | What if there's a problem? |
|---|---|---|
| Whether the entry is complete | Compared with the original PDF catalog page line by line, the first and last chapters are all in | Go back to the original page and re-select when you are out of line, or make up the record after redoing OCR |
| Is the level correct? | There is a difference in indentation or numbering between primary and secondary headings. | Manually unify the hierarchy, do not rely solely on font size |
| Whether the page number corresponds to each other | Select 3 entries at random, open PDF and skip to the corresponding page number to check | The page number dislocation is mostly caused by the page header or page number offset when it is copied, which is corrected according to the original text. |
| Identify whether there is an error | Check numbers, punctuation and proper nouns against the original picture | Both OCR and character recognition can make mistakes, and important contents must be checked manually. |
| Whether to string into the text | Check the results to see if there are any paragraphs outside the catalog | Delete redundant content and narrow the selection next time |
# frequently asked questions (FAQ)
1. What should I pay attention to when extracting pdf directories for free?
When choosing a free extraction tool, first look at three points: whether the file size limit is sufficient, whether the export result is watermarked, and what to do with the file after upload. Papers, contracts and other sensitive documents, give priority to the way that can be processed locally. Whether to keep the typesetting or not, try to do it with one page as far as possible, and then decide whether to deal with the whole document.
2. What if the pdf directory is copied out of garbled code?
Garbled usually comes from font coding or text layer problems. Some PDF fonts are not completely mapped, and garbled or extra spaced characters will appear when copying. You can first export to Word and then check; if the code is still garbled, the text layer of the page is defective, and use OCR to re-identify the page.
3. There is no catalog on the page, you can only see the chapter in the sidebar, how to extract it?
Bookmarks are not necessarily printed on the page. You can expand bookmarks step by step in the reader, manually record the title and jump the page number; when there are many items, you can read the bookmark structure programmatically, output it to text or table, and then check it manually.
4. The page number is misplaced after the catalogue of the scanned piece is identified. What's wrong with it?
The common reason is that the title and page number are split into two lines, or the ellipsis points in the catalog are identified into different symbols. You can correct it one by one against the original page, and don't just rely on the recognition results.
# Summary
When extracting the PDF directory, first determine whether it is selectable text, picture, or sidebar bookmark, and then choose copy, export, or OCR post-processing. After the export, it is checked item by item according to the level, page number and missing lines, and a few articles are selected to check the original text, so that it can be safely used for follow-up writing and arrangement.
