vibioh / fibr Goto Github PK

any way to define landing page? for user/pass no access? NOTE: of course is not an impoprtant request but i ask for if already there are a workaround

setup as:

./fibr_linux_amd64 \
  -publicURL "http://192.168.1.10:10900"  -port 10900 \
  -fsDirectory /home/data  -frameOptions "*" \
  -authProfiles "1:admin" -authUsers "1:$(htpasswd -nBb admin password1)" \
  -thumbnailVideoURL "http://192.168.1.10:10901" -thumbnailImageURL "" \

log:

2021-05-11T17:07:06-04:00 ERROR empty authorization content
2021-05-11T17:07:14-04:00 ERROR invalid credentials

login window popupt with basic auth but does not work, i ask:

QUESTION: using -noAtuh i can setup my own basic authentication? of course all the path will be under that.. but i can defined for specific users?

QUESTION: readme does not xplain how to manage profiles, i only can set profiles when share a file or directorio so that's why i ask the first question!

how to increase the log ? for ? i try to recompile but takes so long time and i dont have good networking

thanks

Dependabot encountered the following error when parsing your .dependabot/config.yml:

The property '#/update_configs/0/automerged_updates/0' contains additional properties ["dependency_type", "update_type"] outside of the schema when none are allowed
The property '#/update_configs/0/automerged_updates/0/match' of type null did not match the following type: object. Is the following line indented correctly?

Please update the config file to conform with Dependabot's specification using our docs and online validator.

checked and try to compile seems need networking and its imposible to package and distribute as is for linux distributions in alpine and debian

also i need to deploy in no docker style.. i mean in a real service.. not as container

since feature of embebed fs templates of commit 21b9b2b now two problems comes:

1. impossible to customized cos all the interface html are embebed
2. golang raized to 1.16 for build

so well its impossible to perform a customized interface ..

Would it be possible for an indepth config process for the readme?

i check that in code all subpath are directories in the publicURL so tried for some option using publicURL and address option and chek tha tall the assets are rquesting from root.

so a new option "--prefixurl "subpath"" can be provided to use a prefix in all path ,

by example http://fbr.dom.fr will become http://fbr.dom.fr/subpath and we can server subpath then!

i made a dirty trick using a symlink in dataroot pointing to "." and it works, but rest of things of course not!

there's ony few releases, limits to amd64 and arm, but those hardware are industri dominated.. not so performance ones like kfreebsd, etc etc or other archs..

due that i need that i posted partially in issue #97 thanks for binary deploy, great!, but is limited to amd64 and arm, there's a wide range of hardware in the market share so a build instructions (including how to put vendos without the traditional go get) is necessary to made packages by me for others arches.. thanks in advance!

i made a build but was so rudimentary .. i will try to fix some things before post feedback here.. that's why i asked you for doc in that topic! i have several x32 machines and i386 ones, cos it consumes less memory and runs in any place.. (all the amd64 are 32bits also) so are cheap to deploy..

Dependabot encountered the following error when parsing your .dependabot/config.yml:

The property '#/update_configs/0/package_manager' value "go" did not match one of the following values: javascript, ruby:bundler, php:composer, java:maven, elixir:hex, rust:cargo, java:gradle, dotnet:nuget, go:dep, go:modules, docker, elm, submodules, github_actions, python, terraform

Please update the config file to conform with Dependabot's specification using our docs and online validator.

errr @ViBiOh the commit 368580d over the issue #146 improves the publicurl handler but only for full domain hijaking, not for subpath, by example:

1. running as : fibr_linux_amd64 -csp '' -noAuth -address "127.0.0.1" -publicURL "http://domain/path1" -port 10900 -fsDirectory /home/data/path1 -frameOptions "*" cause a loop redirection cos the directory are same as the subpath

2. running as: fibr_linux_amd64 -csp '' -noAuth -address "127.0.0.1" -publicURL "http://domain/path1" -port 10900 -fsDirectory /home/data/ -frameOptions "*" do not ause loop redirection.. cos the subpath is a directory in the data dir, but any other links prefix the "path1" so no files working there

3. running as: fibr_linux_amd64 -noAuth -address "127.0.0.1" -publicURL "http://sub.domain" -port 10900 -fsDirectory /home/data/ do not do not work without csp and frameOptions wehen you use any cross domain path cos browser detect the reverse proxy redirection and is not the same domain if the reverse proxy are from main "domain" ("sub,.domain")

all this commit has nothing to do with issue #146

I guess this must be cause of security layer,

I opened fibr server in termux,(phone A). my second (phone B)was connected with it by hotspots,

configuration

fibr  -address 0.0.0.0 -port 8081 -noAuth   -templates "/data/data/com.termux/files/home/fibr/templates/" -fsDirectory "/data/data/com.termux/files/home/fibr_store" -csp "default-src 'self'; base-uri 'self'; script-src 'unsafe-inline'; style-src 'unsafe-inline'; img-src 'self' data:"

I browse http://192.168.43.1:8081 , and able to download Content, but I can't upload with phone B

When I browse http://192.168.43.1:8081 , same , I can't upload with phone A also

But I can file with only localhost, http://localhost:8081

I think , must be any config issue,

what means "multiple backends" in README ? please..

(i hope means any RDBMS for users or similar, not only plain one)

auth Profiles and Users are based on HTTP/Authentication basic auth.. but must parse in each run to the binary.. is there a way to use a .htpass file for?

Newbie here. I wonder if the web listing (especially images/videos) support "infinite scrolling back & forth"?

Web browsers choke if there are more than ~7 pages of picture content to be scrolled through. As a "solution" (actually a nasty workaround but widely adopted - incl. Google, React, etc.) is to completely remove all invisible DOM elements from the DOM and instead put there one and only one block element of the same dimensions as all the removed elements together.

after running new versio got too many logs about "dial tcp: lookup image on 10.1.193.13:53: no such host" so i search on selft binary and got this options :

  -thumbnailImageURL string
    	[thumbnail] Imaginary URL {FIBR_THUMBNAIL_IMAGE_URL} (default "http://image:9000")
  -thumbnailVideoURL string
    	[thumbnail] Video Thumbnail URL {FIBR_THUMBNAIL_VIDEO_URL} (default "http://video:1080")

but no documentation about. any help to propperty configured to disable it?

Hey ViBiOh,
Thanks for this amazing file browser and manager which I have installed and am liking it.

But,it has no copy,cut,paste,rename . Can you add these features to the file manager?

Another feature that I would like to have is downloading the file from the URL . If this is installed in multiuser environment and ssh access to each user is not given and user needs to download file from URL,it is desirable feature to download files from url within file manager.

I would recommend you to take look at grab for implementing downloader feature.

I wonder if there is any easy (even if it includes writing a few lines of Go and subsequent recompilation, I consider it easy 😉) way to allow users with proper rights to "tag" any files (incl. photos & videos) to allow for "generate share link from a search query consisting of this set of tags".

If one has too many files, then sharing with (multiple) others is not any more a hierarchical problem (not 1:many mapping) and thus needs many:many mapping. Therefore I seek something like tagging.

And by tagging I do not mean anything advanced (no DB of tags, no "management of tags" themself, nor anything similar). But just "associate this arbitrary string with this file" (if we had only image files, we could add it to EXIF or similar container; but I would prefer if tagging was not filetype-specific and thus EXIF is not the way to go).

And then "list only files having this set of tags" functionality (including negations "list only files having this set of tags while not having any from this other set of tags" and case insensitive substring matching by default). And persist such queries as "virtual" folders. Then we can share such lists in the same way as regular folders are being shared now.

Any pointers?

Of course, the best would be to have this functionality built-in 😉.