Is it possible to pass a buffer into the converter?

Downloading dependency "pdfbox-app-2.0.16.jar"and "tika-app-1.22" is too slow.
In spite of I try taking these .jar file into "node_modules\pdf2html\vendor ", but it always automatic download.
I don't know how to solve this problem~

Hi,

After converting the pdf to HTML, am getting the same HTML code against all the files(different) and with almost blank body data.
jpg2pdf.pdf

Result HTML:

Hi,
I'm getting this error :
.../[email protected]/node_modules/pdf2html postinstall: throw new Error(Failed downloading dependency ${filename}.); .../[email protected]/node_modules/pdf2html postinstall: ^ .../[email protected]/node_modules/pdf2html postinstall: Error: Failed downloading dependency tika-app-2.6.0.jar. .../[email protected]/node_modules/pdf2html postinstall: at ClientRequest.<anonymous> (/builds/infrastructure/applications_slack/patchs_data/millenium/node_modules/.pnpm/[email protected]/node_modules/pdf2html/postinstall.js:27:23) .../[email protected]/node_modules/pdf2html postinstall: at ClientRequest.emit (node:events:518:28) .../[email protected]/node_modules/pdf2html postinstall: at TLSSocket.socketErrorListener (node:_http_client:495:9) .../[email protected]/node_modules/pdf2html postinstall: at TLSSocket.emit (node:events:518:28) .../[email protected]/node_modules/pdf2html postinstall: at emitErrorNT (node:internal/streams/destroy:169:8) .../[email protected]/node_modules/pdf2html postinstall: at emitErrorCloseNT (node:internal/streams/destroy:128:3) .../[email protected]/node_modules/pdf2html postinstall: at process.processTicksAndRejections (node:internal/process/task_queues:82:21) .../[email protected]/node_modules/pdf2html postinstall: Node.js v20.11.0 .../[email protected]/node_modules/pdf2html postinstall: Failed
Anyone having the same ?
Thank you

This library uses the fs library, which doesn't exist on next.js frontends. Why would this library need the file system in the first place? Getting/setting from the file system might not work in serverless environments.

Might need to refactor to not use the fs library or any other library not available to the client or in a serverless environment.

I've started up a nodejs project and installed pdf2html?. However, whenever I try to run pdf2html.html()` i get the following error message:

java -jar /Users/<path to project>/node_modules/pdf2html/vendor/tika-app-1.22.jar --html ./mock/example.pdf
Oct 08, 2020 10:27:51 AM org.apache.tika.config.InitializableProblemHandler$3 handleInitializableProblem
WARNING: J2KImageReader not loaded. JPEG2000 files will not be processed.
See https://pdfbox.apache.org/2.0/dependencies.html#jai-image-io
for optional dependencies.

Oct 08, 2020 10:27:51 AM org.apache.tika.config.InitializableProblemHandler$3 handleInitializableProblem
WARNING: org.xerial's sqlite-jdbc is not loaded.
Please provide the jar on your classpath to parse sqlite files.
See tika-parsers/pom.xml for the correct version.
Exception in thread "main" java.net.MalformedURLException: no protocol: ./mock/example.pdf
        at java.base/java.net.URL.<init>(URL.java:668)
        at java.base/java.net.URL.<init>(URL.java:564)
        at java.base/java.net.URL.<init>(URL.java:511)
        at org.apache.tika.cli.TikaCLI.process(TikaCLI.java:488)
        at org.apache.tika.cli.TikaCLI.main(TikaCLI.java:149)

How do I solve this?

To upgrade log4j version 2.15.0 to 2.17.0
Ref https://dist.apache.org/repos/dist/release/tika/2.2.1/CHANGES-2.2.1.txt

Hello.
A new Tika version was released and the endpoint 'https://dlcdn.apache.org/tika/2.2.1/tika-app-2.2.1.jar' that the postinstall tries to access now returns a 404 error, because of this it is not being possible to install the package. Please update the dependency to use the new version 2.3.0.

Hi,

When you have to use a HTTP proxy to make web requests, the postinstall script will fail. This is because is uses http.get which does not handle HTTP_PROXY environment variable automatically.

One option to fix this would be to use http-proxy-agent, cf. https://github.com/ambanum/CGUs/blob/master/src/fetcher/index.js#L9-L14.

Another option, without modifying the code, would be to provide infos in the README file about this particular setup and getting pdf2html to work. This basically boils down to:

cd node_modules/pdf2html/vendor
# These URLs come from https://github.com/shebinleo/pdf2html/blob/master/postinstall.js#L6-L7
wget http://archive.apache.org/dist/pdfbox/2.0.16/pdfbox-app-2.0.16.jar
wget http://archive.apache.org/dist/tika/tika-app-1.22.jar

Thanks

When I tried to convert a PDF that contained MALAYALAM language the output was not as expected.

Input PDF Content

എം എൽ എ യുടെ അധ്യക്ഷതയിൽ യോഗം

Output

���ി�ിൽ എം എൽ എ �െട ആധ��തയിൽ േയാഗം

I am converting quite a big file and I encounter the following error:

     Conversion error: RangeError [ERR_CHILD_PROCESS_STDIO_MAXBUFFER]: stdout maxBuffer length exceeded

Can anyone look into the problem and help me, please?

We have been seeing a lot of issues related to Tika app JAR (404), can we somehow commit the jar into the repository instead of download it directly from apache.org?

Any thoughts?

Regarding this in the docs, "Java runtime environment (JRE) is required to run this module."

Next.js and/or serverless backends are very popular right now. However, it might be very hard or impossible to set up JRE in Next.js, AWS Lambda, etc.

Hello~Appreciate your great work on this.

I'm using this package, but the postinstall script always fails for me, due to the slow network and firewall limits.

And the https module this package is using doesn't support the HTTP_PROXY environment variable. This is a headache for me...

To resolve the above issue, we might have two solutions:

1. Download requests honor the HTTP_PROXY, HTTPS_PROXY environment variables. We can use another request library that supports so.

2. Allow users to specify download URLs. So that users can specify a binary download URL that is served by a mirror that is located close to clients.

For this solution, we can borrow how node-sass support this for users from worldwide:

https://github.com/sass/node-sass/blob/24741b351cb046c4548e77886647cd4c89b48c66/lib/extensions.js#L192-L199

I would like to use buffer of a file except url.

eg.

const bufferPdf = Buffer.from(pdfToSave.buffer);

      await pdf2html
        .text(pdfBuffer, (err, html) => {
          if (err) {
            console.error("Conversion error: " + err);
          } else {
            console.log(html);
          }
        })
        .promise();

The installation url for tika-app-2.3.0.jar does not exist, causing an error in the package installation.

is there any way to get pdf (Blob) to Html using
pdf2Html
as of now pdf2Html needs URL of Pdf file to convert it as HTML
but I have a Pdf content or Blob so how to make use of it, is there any hidden properties in it
to get it used

while attempting to npm install pdf2html, i'm getting:

Error: Failed downloading dependency tika-app-2.1.0.jar.

and/or

Error: Failed downloading dependency pdfbox-app-2.0.24.jar.

Depending on the project. I'm grasping at straws here and wondering if it has to do with the recent log4j issues, but figured i'd at least leave the report before diving deeper.

Update: indeed pdfbox-app-2.0.24.jar is no longer available on the site, but 2.0.25.jar is

Hi,

https://github.com/shebinleo/pdf2html/blame/79f4eff672ad924094acf971687ecf75f108385b/index.js#L22

By not listening to this event, the stderr filled up with warnings about fonts that were not installed and, according to this issue, this is why my processing stopped, the 'close' event was never sent.

Could you possibly listen to the event again or be able to pass an option for the stdio param? This has solved the issue on my system.

Thanks!

This is a fresh project with nothing in it other than Next.js and this library. I tried out the example code in the README and got an extremely cryptic error.

Code:

import { NextApiRequest, NextApiResponse } from "next";
import pdf2html from "pdf2html";

export default async function pdfToHtml (req: NextApiRequest, res: NextApiResponse) {
    
  const html = await pdf2html.html("/pdfs/smallPdfExample.pdf")
  console.log("html: ", html)

  res.status(200).json({ name: 'John Doe' })
}

Error:

Hi, pdf2html.pages & pdf2html.html only returns paragraphs & links, but no images. Is this possible? Are there any workarounds available to get images (in the right order between other textual elements)?

when i use through follow in project created by create-react-app,throw error:

const pdf2html = require('pdf2html')
 
pdf2html.html('sample.pdf', (err, html) => {
    if (err) {
        console.error('Conversion error: ' + err)
    } else {
        console.log(html)
    }
})

Hello, I do I make sure that the text formatting, such as bold, italic, bigger font size, such formatting are preserved when I am converting the file to html.

error: Command failed: java -jar D:\vscodeProject\node\myserve\node_modules\pdf2html\vendor\tika-app-2.4.0.jar --html ../docs/9.5-9.17.pdf
Error: Registry key 'Software\JavaSoft\Java Runtime Environment'\CurrentVersion'
has value '1.8', but '1.7' is required.
Error: could not find java.dll
Error: Could not find Java SE Runtime Environment.

code is like;

const options = { text: true }

pdf2html.pages('m.pdf',options,(err, htmlPages) => {
    if (err) {
        console.error('Conversion error: ' + err)
    } else {
        console.log(htmlPages)
    }
})

pdf is like

Name: __ZZH_____

but find text is Name: __________(Ps; it is a input box)
Can anyone Help the case