How to edit the extracted document content online and retain the layout

Key points of this article

If the extraction result is a Word file, modify it directly in the web editor, and the layout is usually relatively complete; if it is a PDF, check the text first and then convert it to Word for editing.
The text in the scanned document must first undergo OCR recognition, and the recognition results must be checked item by item with the original. Pay special attention to numbers, dates, and proper nouns.
Modify paragraph by paragraph in the web editor. Edit tables and formulas in their original positions. Do not retype the entire paragraph. This will make it easier to maintain the original layout.
Before exporting, check the title level, table row and column numbers, and formulas against the original. If any problems are found, go back to the corresponding steps to correct them. Do not distribute directly.
Use a shared link to send when multiple people are required to collaborate; choose the export format according to subsequent use, use Word for submission or handing over to colleagues, and use HTML for publishing web pages.

After extracting text and tables from a contract, a paper or a report, many people find that the content is usable, but the layout is completely messed up: the title level is lost, the tables are misplaced, and the formulas become garbled. The problem often lies in the step after "extraction", that is, how to modify the content without destroying its original appearance.

This article is written for those who need to continue to modify the document after extraction, such as readers who want to rewrite a report, update a data table, or supplement a paragraph of a paper. After reading, you will know how to choose the editing method according to the file format, how to operate step by step in the web editor, and what to check before exporting.

How to edit the extracted content online so that the layout is not messy? One sentence conclusion

The format of the extracted content determines where you start editing. The Word format is suitable for direct entry into a web page editor; the layout of PDF is fixed according to position, and direct modification can easily cause extrusion and misalignment, so it usually needs to be converted to Word first. The table below can help you make a quick judgment.

your situation	Recommended practices	reason
The extraction result is a Word file (.docx)	Edit directly after uploading app.unidoc.top	The web page editor is compatible with Word editing, and paragraph and table structures are more completely preserved.
The extracted result is a PDF, and I only want to change some text.	First use the desktop PDF editor to change the paragraph text, and then export it to Word	Text in PDF is typed according to position. Correcting text in PDF first can reduce subsequent misalignment.
There are formulas in the content	Modify the formula in the original position in the web page editor and check again after modification.	Formula symbols, superscripts and subscripts are prone to errors during conversion and need to be confirmed one by one.
Text extracted from scanned document	Do OCR recognition first, then enter editing	The scanned document does not have a text layer that can be selected and can only be modified after recognition.
After modification, send it to colleagues or put it on the web page	Send with a share link, or export to HTML on demand	Sharing and exporting can make the editing results available to others.

Before you start: Confirm these points first

Before starting, take a few minutes to judge the following things, which can save most of the rework.

Confirm file format. Open the extraction result to see if it is Word, PDF, or other formats. The format is different, the path is different, and the following steps will be different.
Determine whether it is a scanned copy. Try selecting a piece of text with the mouse and copy it. If you can select and copy the correct text, it means there is a text layer in the file; if you cannot select or copy it with garbled characters, it is usually a scanned document and requires OCR first.
Keep an original copy. Paragraphs may be accidentally deleted or formatting may be overwritten during online editing, so keep the original for item-by-item comparison.
Clear purpose. Should I hand it over to colleagues for revision, submission, or publish it as a web page? The purpose determines whether to export to Word, HTML, or LaTeX.
Prepare a checklist. Numbers, dates, names, and professional terms are easy to make mistakes. List them first and then check them one by one.

Modify the steps to extract content in UniDoc web editor

The following takes a 30-page paper draft extracted from PDF and converted into Word as an example to explain how to complete the modification in the web version.

Prepare Word file. If the original extraction result is a PDF, first check and correct the text in the PDF editor of the desktop client, and then export to Word.
Open the web editor. Visit app.unidoc.top in your browser, which is the address of the UniDoc web editor.
Upload the Word file. Import the prepared .docx file into the editor, wait for the content to be fully displayed, and then read it through to make sure there are no obvious missing paragraphs or tables.
Revise the text paragraph by paragraph. Make changes in sequence by paragraph. After each paragraph is modified, check whether the title level and paragraph spacing are still consistent with the original. Do not check them together at the end.
Edit tables and formulas in their original location. Change the numbers in the table directly in the cells, do not delete the entire table and re-enter it; the formulas are also modified in the original position to avoid damaging the alignment.
Check the layout and save. Scroll through the full text to see if the paging and paragraphs are normal, and save the current version after confirming that they are correct.
Export or share according to usage. Export Word when you need to leave it to your colleagues to continue editing; export HTML when you want to publish it as a web page, and the web version supports lossless export; when you need to continue writing in the LaTeX environment, you can use the round-trip method between Word and LaTeX. If you only want others to view or comment, you can generate a sharing link.

Check each item after completion

Once you get the results, don’t distribute them directly. Check item by item against the original, and if any problems are found, go back to the corresponding steps to correct them.

Check item	What do you think?	What if there's a problem?
Title level	Check whether the font size and position of the first-and second-level titles are the same against the original.	Reset the level in the corresponding paragraph, and then scroll through the full text to confirm
Number of rows and rows in the table	Count rows and columns, compare them with the original, especially merge cells	Go back to the modification step, complete or delete it in its original location, and don't re-paste the entire table.
Formula	Confirm symbols, superscripts, fractions, and parentheses one by one	Correct in the original position, export after correction, and proofread it with the original.
Numbers, dates and proper nouns	Compared with the original item by item, the content of OCR identification should be focused on.	Correct directly, check back the source, make sure it is correct, and then save it
Paging and paragraph	Scroll through before export to see if there are any solitary rows, broken tables or misplaced pictures and text.	Re-check the adjacent pages after adjusting the paragraphs to ensure the connection between the front and back

It should be reminded that if the content has been OCR or machine translated, be sure to check the important paragraphs manually, not just rely on automatic results.

# frequently asked questions (FAQ)

1. Can the content extracted from the scanned piece be changed directly in the web editor?

Usually it cannot be changed directly, and there is no editable text layer for the scanned piece. First use OCR to identify the text, and then enter the editor, the recognition results should be checked item by item with the original.

2. Will online editors lose their original format?

The page editor is compatible with Word editing and retains most of the paragraph and table structure. However, there may still be slight changes in complex typesetting, and it is safe to check against the original before export.

3. Why is the layout messed up when the PDF is changed directly?

PDF text is typeset according to fixed coordinates, adding or deleting a few words will squeeze the following content. It is recommended that you first convert to Word and then modify it in the editor.

4. What is the difference between the online web version of the document content extraction and the desktop side?

The web version of UniDoc focuses on Word compatible editing, formula and multi-format export; the desktop client also includes a PDF editor that can modify PDF paragraph text, do OCR and export Word. The two can be used together.

5. Can I change the edited content with my colleagues?

You can use the sharing link to send the content to others. For specific viewing or editing permissions, please refer to the actual interface.

# Summary

If the extracted content wants to be edited online without random layout, the key is to choose the path according to the format: Word directly enters the web editor, PDF modifies the text before conversion, and the scanned part does OCR first. When the revision is carried out paragraph by paragraph, edited in the original position, and checked against the original item by item before export, the dislocation and omission can be controlled within the controllable range.

Open OmniDoc and export the PDF content as an editable file