Instead of building the binary from master it would be good to use the tags. It probably would be more stable. Current newest version is 6.0.3.

Could you please remove --disable-pdfimport from build script? With this flag, lo cannot convert PDF to other formats such as doc, which should be a very common use case.

I was able to compile my own binary without it, but it would be better if your release already contains pdfimport. :)

Thanks!

I'm using the latest version of libre office. Some doc files are unable to convert to pdf files. Is there any bug identified already and planning to release a new version?
Note: When I'm trying to open the doc file I'm able to open the file. File is not corrupted but it is not getting converted to pdf.

Hello, I'd like to port your lambda configuration from terraform to serverless, but I don't know terraform. Is there a chance that you could provide some details of what the different terraform files are meant for? Thanks for this cool project! 👍

Updating my AWS Lambda function with the new LibraOffice compiled for Amazon Linux 2 and when I run the command it throws com::sun::star::container::NoSuchElementException'.

I've confirmed the file is there and tried both pptx and docx but it throws the same exception. I did notice in a test that you run the command twice with it expecting to fail the first time, not sure if it's related.

Thanks for your function and work on this, very much appreciated.

I'm trying to create a aws lambda docker image with the lo.tar.gz pre-unzipped inside the image using the following dockerfile:

FROM public.ecr.aws/lambda/python:3.8

RUN yum install -y tar
RUN yum install -y gzip
RUN yum install -y wget

RUN wget https://github.com/vladgolubev/serverless-libreoffice/releases/download/v6.4.0.1/lo.tar.gz
RUN tar -xf lo.tar.gz --verbose

I get the following error from the docker daemon when building this dockerfile:

Error response from daemon: Error processing tar file(duplicates of file paths not supported): 
Failed to deploy '<unknown> Dockerfile: ../Dockerfile': Can't retrieve image ID from build stream

I did not get this error with previous versions of this serverless-libreoffice package, only on the python 3.8 supported version.

Any ideas what goes wrong? I've also tried running this with a different base image and i get the same error.

Having Libreoffice as a Layer inside Lambda would be even more awesome.

Seems the python example no longer works that is located here:

https://github.com/vladgolubev/serverless-libreoffice/blob/master/STEP_BY_STEP.md

Seems tar/zip is no longer available in amazon linux 2 python3.8

Ive also attempted to make it a layer, but I get this error when running it:

sh: /opt/instdir/program/soffice: No such file or directory

Even though it clearly is there.

Anyone else get this working?

First of all - great job on this. I'm enjoying playing around with it.

There was one small hick-up when creating everything with terraform ( the aws_iam_role was created, but it didn't have the aws_iam_role_policy attached to it yet before the lambda was created - as it depends on the lambda function arn. And the lambda function could not be created, because the role did not have a aws_iam_role_policy assigned yet, and lacked the xray:PutTraceSegments and xray:PutTelemetryRecords permission yet ).

I'm not sure if this was an issue with a recent version, but I fixed it by changing the resource limitation in the aws_iam_policy_document to * instead of ${aws_cloudwatch_log_group.convert.arn}. Maybe that's how you started, and then you limited the resources later on?

Now to the issue :
It would be nice to split the .zip file from the code - so when working on the lambda code, you don't have to keep uploading the 140-150mb image with it every time. Maybe storing it in s3, or in a different folder would suffice? ( https://stackoverflow.com/a/35188706 )

Has anyone had success transforming the lambda layer from python 3.8 to python lambda 3.9?

Thank you, this is great repo:)

So, I am running with lambda runtime of python 3.8.0 but I got the following error.

Fontconfig error: Cannot load default config file

Is there any solution?

First of all let me say that this is a great project and and I have managed to get this working on an independent machine.

However, in the context of the Lambda function it doesn't seem to work. The output does not get created. The input file size about 100kb and I have allocated the recommended RAM and the timeout values. Following is the error I get:
{
"StatusCode": 200,
"FunctionError": "Handled"
}
{"errorMessage":"ENOENT: no such file or directory, open '/tmp/${outputFilename}'","errorType":"Error","stackTrace":["Object.fs.openSync (fs.js:646:18)","fs.readFileSync (fs.js:551:33)","exports.handler (/var/task/code.js:19:28)","","process._tickDomainCallback (internal/process/next_tick.js:228:7)"]}

Any pointers or help is greatly appreciated.

Hi,
First of all thanks for sharing this awesome module.
I am trying to use the complied file but getting the error added on subject.
I have followed the doc step by step.
I uploaded the lo.tar.gz on S3 bucket and created a lambda function with code which is added by you.
I am getting following error:

tar: This does not look like a tar archive

gzip: stdin: not in gzip format
tar: Child returned status 1
tar: Error is not recoverable: exiting now
START RequestId: 5065489b-7b82-11e8-9818-73a78f241cd9 Version: $LATEST
module initialization error: Error
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed

0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0
100 243 0 243 0 0 224 0 --:--:-- 0:00:01 --:--:-- 224
tar: This does not look like a tar archive

gzip: stdin: not in gzip format
tar: Child returned status 1
tar: Error is not recoverable: exiting now

Hi, how do I compile it to support non-English languages?

I'm trying to convert excel files to PDFs and it is successfully completing this action. The issue is none of the columns are expanded so there is a lot of data that is missing in the pdf form. Is there some option that can be added to the convert command to force these columns to expand? Is there some other way to get the columns to expand so all of the data is available in the pdf converted file?

In result of conversion of any .msg file I get incorrect pdf - it contains symbols ���
May be it because err:"Fontconfig error: Cannot load default config file" during conversion

Hello, I am having problems when converting files to pdf in lambda using the eu-west-1 layer (https://github.com/shelfio/libreoffice-lambda-layer, tried both gzip and brotli options). I am using the 3.0.0 version of @shelf/aws-lambda-libreoffice.
I use AWS SAM to deploy the lambda to AWS and the strange thing is that there are no problems locally.
This is the error that I get when trying to convert a file:

{
  "errorType": "Error",
  "errorMessage": "Command failed: cd /tmp && /tmp/instdir/program/soffice.bin --headless --invisible --nodefault --view --nolockcheck --nologo --norestore --nofirststartwizard --convert-to pdf --outdir /tmp /tmp/dog.png\n/tmp/instdir/program/soffice.bin: error while loading shared libraries: libmergedlo.so: cannot open shared object file: No such file or directory\n",
  "trace": [
    "Error: Command failed: cd /tmp && /tmp/instdir/program/soffice.bin --headless --invisible --nodefault --view --nolockcheck --nologo --norestore --nofirststartwizard --convert-to pdf --outdir /tmp /tmp/dog.png",
    "/tmp/instdir/program/soffice.bin: error while loading shared libraries: libmergedlo.so: cannot open shared object file: No such file or directory",
    "",
    "    at checkExecSyncError (child_process.js:629:11)",
    "    at execSync (child_process.js:666:13)",
    "    at convertTo (/var/task/node_modules/@shelf/aws-lambda-libreoffice/lib/convert.js:39:40)",
    "    at process._tickCallback (internal/process/next_tick.js:68:7)"
  ]
}

Is it possible to update the compile.sh to match the release for: 6.4.0.1 It seems to be still for 6.2.1.2
I can download the generated archive, but it would be great to be able to replicate the compilation process.

I'm trying to get this working on Python 3.8, however I'm not having any luck. With the version I compile I get SalAbort: 'Unknown application error' (after resolving a lot of missing shared libs). I'm using this AMI to compile on: https://ap-southeast-2.console.aws.amazon.com/ec2/v2/home?region=ap-southeast-2#Images:visibility=public-images;search=ami-06ce513624b435a22;sort=name

Also when I download the release from this repo for 6.4.0.1 I'm getting the following error when I try to run my Lambda function: terminate called after throwing an instance of 'com::sun::star::container::NoSuchElementException'

I'm only getting the errors when I try to run the LibreOffice Binary in Lambda when I test it on an EC2 instance running the above AMI everything works fine.

It appears that the gcc72-c++ package now has a NEWER version of libstdc than is really used in the lambda environment. The lambda environment has libstdc++.so.6.0.20 and the latest version of gcc72-c++ has .24. Your scripting worked well, but the resulting soffice.bin wouldn't run in lambda because of this. Easiest solution was to just take the .24 version of the library and plunk it into instdir/program and then creating a soft link in there as well called libstdc++.so.6 that points to that now-local .24 version. Might need to add that to the scripting. Other option would be to specify a hard version of gcc72-c++ since as of 3/8/19 it's using libstdc++72-7.2.1-2.59.amzn1.x86_64. A slightly older version there probably is what includes the .20 version of libstdc. Thanks for all the hard work.

NodeJS 8.10 is EOL from Feb 3 and NodeJS 10.x uses Amazon Linux 2. I've tried to compile it in this AMI doesn't seem to work and fails with:

strip: './instdir/**/*': No such file

At the moment it's likely I'll need to move to Python however 3.8 is also on Amazon Linux 2 and by the looks of this it will be the same issue as this is the error I get when trying to run the current compiled version in NodeJS 10.x.

Thanks a heap for this work, very much appreciated.

I use nodejs10 aws lambda + lambda layer created from the latest lo.tar.zip.
It works well, but conversion Word docs in which "Table of Contents" exists, raises an error: "Please verify input parameters... (SfxBaseModel::impl_store file:///tmp/file.pdf failed: 0xc10(Error Area:Io Class:Write Code:16))"
With previous version - lo.tar.gz the same document is converted successfully

Run command

aws lambda invoke --function-name convert-to-pdf \
>     --payload '{"filename":"mr.xlsx"}' output.txt && cat output.txt

and got the answer

{
    "FunctionError": "Unhandled",
    "ExecutedVersion": "$LATEST",
    "StatusCode": 200
}
{"errorMessage": "An error occurred (403) when calling the HeadObject operation: Forbidden", "errorType": "ClientError", "stackTrace": [["/var/task/lambda_function.py", 12, "lambda_handler", "s3_bucket.download_fileobj(inputFileName, data)"], ["/var/runtime/boto3/s3/inject.py", 720, "bucket_download_fileobj", "Callback=Callback, Config=Config)"], ["/var/runtime/boto3/s3/inject.py", 678, "download_fileobj", "return future.result()"], ["/var/runtime/s3transfer/futures.py", 73, "result", "return self._coordinator.result()"], ["/var/runtime/s3transfer/futures.py", 233, "result", "raise self._exception"], ["/var/runtime/s3transfer/tasks.py", 255, "_main", "self._submit(transfer_future=transfer_future, **kwargs)"], ["/var/runtime/s3transfer/download.py", 353, "_submit", "**transfer_future.meta.call_args.extra_args"], ["/var/runtime/botocore/client.py", 314, "_api_call", "return self._make_api_call(operation_name, kwargs)"], ["/var/runtime/botocore/client.py", 612, "_make_api_call", "raise error_class(parsed_response, operation_name)"]]}

The function was created manually and via CLI. The same problem.

Hi,
First of all thanks for sharing this awesome module.
I am trying to use the complied file but getting the Following error.

/tmp/instdir/program/oosplash: error while loading shared libraries: libX11.so.6: cannot open shared object file: No such file or directory

I have followed the doc step by step.
I an using this on AWS Lambda (Python 3.x)

Can you please advice me, what can i do to resolve this error.

Thanks in advance.

I am using the step by step example to convert a word document to a pdf and i get this error:

Error: Command failed: cd /tmp && /tmp/instdir/program/soffice.bin --headless --invisible --nodefault --nofirststartwizard --nolockcheck --nologo --norestore --convert-to pdf --outdir /tmp "128-performance based template.docx" terminate called after throwing an instance of 'com::sun::star::container::NoSuchElementException'

Could you help me out with what the issue is?

I confirm I have this file in my /tmp folder.

When document containes some WordArt or special font, the result seems unapplicable. Some word and font style was lost and layout was changed.

This issues may be caused by LibreOffice. Are there any method to add own font for each convertion task?

The file downloaded from releases tab contains everything. Do you have a trimmed .zip that can be uploaded direct to Lambda?

Thanks!

Hello
A couple mistakes on converting html->pdf:
I need convert msg files to pdf. As LibreOffice can't convert msg to pdf directly, i do pre-conversion msg-> html by another app. And AWS Lambda with LibreOffice converts this html to pdf.
Result pdf contains empty, extra first blank page and doesn't contain "From" email part. I've tried desktop LibreOffice app, it converts the same html correctly.
Could you please advise?

Hi there.

I'm experiencing differences in the output pdf format (converting from docx) when running serverless-libreoffice vs running libreoffice 5.4 pre-installed on the machine.

Namely, it seems that certain text is varying in size, some being bolded while others are not. I assume this has something to do with fonts, but can you think of any way to rectify?

I've attached the three files, the original docx that was used, the Locally created pdf and also the Lambda created pdf.

sample-docx.docx
lambda.pdf
local.pdf

Thank you.

I'm trying to set this up in Azure functions, a python function which runs on Linux. I've managed to do some of the steps, but am now stuck on the conversion command.

Actually, what I need is to convert EMF and WMF files to PNG - not the entire PDF converter. But I'm trying to set up the PDF converter from a simple .TXT file first.

I've so far managed to:

1. Connect to Azure Storage (S3 bucket equivalent) and pull the "test.txt" file into a "/tmp" folder.
2. (I push the lo.tar.gz file along with the code - so no need to pull it from Storage/Bucket) Run the "tar -xf lo.tar.gz" and confirm that I get the executable at "/instdir/program/soffice.bin"
3. Execute "instdir/program/soffice ~~" using subprocess (rather than os.system, so that I get the output and error messages) However, I get an error here:

        STDOUT: '',
        STDERR: '/tmp/functions\\\\standby\\\\wwwroot/instdir/program/soffice.bin: error while loading shared libraries: libssl3.so: cannot open shared object file: No such file or directory\\n',

I'm really inexperienced with Linux - so I don't understand the error message I'm getting here. Does it mean there's some system files that are required, which aren't included in the lo.tar.gz file?

I have used the lo.tar.gz file you had precompiled for Lambda. Do you reckon it should work here as well - or would I need to build my own specific file for use on Azure Functions (Python functions run on Linux)?

Any help here is much appreciated. And of course I'll share the code for Azure Functions once I manage to get it working - if that's interesting.

Firstly just wanted to say that you've done a fantastic job here, this seems like it could lead to a really neat way of handling file conversion tasks without having to pay through the nose for expensive third parties.

I'm currently setting up your scripts to have a play with this for a project idea, I noticed on your site you use an AWS API Gateway. This isn't something I've used before but if my understanding is correct, you can push file uploads directly to that, and presumably you've got a script there that handles pushing on to s3/lambda right? Do you have a sample file for API Gateway available?

When I convert a doc file which file size is 10M, the lambda will time out, do you have any solution about this? And when I convert an excel file which has too many columns, the PDF contents will not correct.

Hi,
I don't want to convert files to pdf, but to docx. But, when I try to do that it fails, saying no filters found for docx. When I pass a filter(writer_MS_Word_2007) to the command, it still fails.
The error is "Error: Please verify input parameters... (SfxBaseModel::impl_store file:///xyz/tmp/hello.docx failed: 0x81a". It is not working on lambda nor in my local. But, pdf conversion is successful.
Does this mean that export filters for docx are missing? When I copied my local installation and ran the python script locally, it is working. Can you please tell me how to fix this?

Hello,
Firstly, thanks for sharing your code, it's a great help.

I'm currently setting up your scripts and more particularly LibreOffice on AWS lambda. However, when I want to execute soffice command to convert a document I have the following error:

_ { Error: Command failed: cd /tmp && ./instdir/program/soffice --headless --invisible --nodefault --nofirststartwizard --nolockcheck --nologo --norestore --convert-to pdf --outdir /tmp/ fileTest

terminate called after throwing an instance of 'com::sun::star::uno::RuntimeException'
terminate called recursively

Fatal exception: Signal 6
Stack:
/tmp/instdir/program/libuno_sal.so.3(+0x3ca00)[0x7ff76c8eba00]
/tmp/instdir/program/libuno_sal.so.3(+0x3cb70)[0x7ff76c8ebb70]
/lib64/libc.so.6(+0x35270)[0x7ff76c520270]
/lib64/libc.so.6(gsignal+0x37)[0x7ff76c5201f7]
/lib64/libc.so.6(abort+0x148)[0x7ff76c5218e8]
/var/lang/lib/libstdc++.so.6(_ZN9__gnu_cxx27__verbose_terminate_handlerEv+0xed)[0x7ff765a8faed]
/var/lang/lib/libstdc++.so.6(+0x5daf6)[0x7ff765a8daf6]
/var/lang/lib/libstdc++.so.6(+0x5c6b9)[0x7ff765a8c6b9]
/var/lang/lib/libstdc++.so.6(__gxx_personality_v0+0xd2)[0x7ff765a8cd82]
/var/lang/lib/libgcc_s.so.1(+0x10070)[0x7ff765829070]
/var/lang/lib/libgcc_s.so.1(_Unwind_Resume+0x59)[0x7ff7658296f9]
/tmp/instdir/program/libmergedlo.so(+0x2ef3cb1)[0x7ff76fa00cb1]
/tmp/instdir/program/libuno_sal.so.3(+0x16c62)[0x7ff76c8c5c62]
/tmp/instdir/program/libuno_sal.so.3(+0x3cb2f)[0x7ff76c8ebb2f]
/lib64/libc.so.6(+0x35270)[0x7ff76c520270]
/lib64/libc.so.6(gsignal+0x37)[0x7ff76c5201f7]
/lib64/libc.so.6(abort+0x148)[0x7ff76c5218e8]
/var/lang/lib/libstdc++.so.6(_ZN9__gnu_cxx27__verbose_terminate_handlerEv+0x15d)[0x7ff765a8fb5d]
/var/lang/lib/libstdc++.so.6(+0x5daf6)[0x7ff765a8daf6]
/var/lang/lib/libstdc++.so.6(+0x5db41)[0x7ff765a8db41]
/var/lang/lib/libstdc++.so.6(__cxa_rethrow+0x46)[0x7ff765a8ddd6]
/tmp/instdir/program/libmergedlo.so(_ZN12SvtSysLocaleC1Ev+0x33b)[0x7ff76f6d56db]
/tmp/instdir/program/libmergedlo.so(+0x1407a7b)[0x7ff76df14a7b]
/tmp/instdir/program/libmergedlo.so(+0x1eb9983)[0x7ff76e9c6983]
/tmp/instdir/program/libmergedlo.so(+0x1ec0359)[0x7ff76e9cd359]
/tmp/instdir/program/libmergedlo.so(+0x2ef54b6)[0x7ff76fa024b6]
/tmp/instdir/program/libmergedlo.so(_Z6SVMainv+0x22)[0x7ff76fa025a2]
/tmp/instdir/program/libmergedlo.so(soffice_main+0x8a)[0x7ff76e9ed54a]
/tmp/instdir/program/soffice.bin[0x4006eb]
/lib64/libc.so.6(_libc_start_main+0xf5)[0x7ff76c50cc05]
/tmp/instdir/program/soffice.bin[0x400721]

Do you have any idea? It seems that the lo.tar.gz contains a binary file that AWS lambda can't open. I tried to do the same thing on MAC OSX, and I have the following error:

./instdir/program/soffice: line 174: AWS/instdir/program/oosplash: cannot execute binary file
./instdir/program/soffice: line 174: AWS/instdir/program/oosplash: Undefined error: 0

I checked and this oosplash is present in the directory, I tried to change the permissions but it didn't work.
By the way, I don't have this issue when I use the standard LibreOffice distribution.