(No comment was entered for this change.)

Well, technically, any executable or file hosted on the web must first be downloaded 
before it could be opened or run.  Once the file is downloading, you can select that 
file (in the bottom) and choose to open/run it automatically upon completion of the 
download.

Or, if it was downloaded previously, you can re-visit your downloads from within 
Chrome and can open/run it from there.

So, I don't see this as being an issue inasmuch as a learning curve.

With IE and Firefox (and Opera) if you choose to Run/Open a file it downloads to your
TEMP directory and opens the file from there.  This way your download directory isn't
cluttered with files you only ever meant to view once.

This is a duplicate of https://code.google.com/p/chromium/issues/detail?id=338

I can understand where people are coming from with downloading it and then running it 
however it's not always necessary to do that. Take, for example, Torrents or those 
web setup programs that the likes of Yahoo! and Microsoft love so much. When you 
launch a torrent, most programs nowadays will copy the file into their own folder, 
leaving you with a redundant file in your download folder. As for programs that use a 
web setup installer, once you've performed the initial run, the file is useless to 
you. If you want to change or repair, you can use the Add/Remove Programs entry. If 
you want to reinstall it, you can either use the cached 'real' installer or go and 
re-download the tiny little boot-strapper. Either way, without this 'run' option, you 
end up with a bunch of 'temporary' files which you'll probably never use again yet 
are cluttering up your download folder.

I can understand that this is a minor feature request and there are way more 
important things out there that need addressing but I feel that this should at least 
be considered as it's more than a request for feature cloning, it's a practical 
function that can be very useful in a number of different scenarios.

(No comment was entered for this change.)

 Issue 338  has been merged into this issue.

I don't know how to do it, but it would be useful to surface a UI equivalent of IE's 
"Run" (without IE's intermediate modal dialog).

I like how it is. just set chrome to not ask where to save. Then when you download a 
file, you just click save, and when it's done downloading, click its icon in the 
download bar to run it.

Yes, but what if I don't want the file SAVED on my computer permentely? I'd rather 
have it save in a temporary folder automaticlly when I click 'Run' when I save it.

Possible solution for you to directly view PDF files in Chrome.
I was looking for solution to this too and in my search I found markmail.org's forum.

Try this...

1. Open Google Chrome.
2. In the address bar, type "about:plugins". You'll see a long list of all of your 
browser's plugins.
3. Do a page search for "Adobe Acrobat." (Search within a page by hitting Ctrl+F).
4. You should see this > File name: nppdf32.dll
5. If not do the following.

If you are on Window XP,

COPY

nppdf32.dll 

FROM

\Program Files\Adobe\[Your installed Acrobat version]\Acrobat\Browser

TO

[username]\Local Settings\Application Data\Google\Chrome\Application\Plugins\


If you are using Window Vista,

COPY

nppdf32.dll 

FROM

\Program Files\Adobe\[Your installed Acrobat version]\Acrobat\Browser 


TO

[username]\AppData\Local\Google\Chrome\Application\Plugins



That should work. 


[Ref URL :http://markmail.org/message/bzypulp6lsmixkur]

This request is here from first public beta. Many people need open/save dialog, same 
like me. Because without this dialog is opening all attachments in email (for 
example) really annoying. Where is problem insert one new dialog between browser and 
"save as" dialog. All browser have it. Why you can't give people this useful choice. 
This dialog is the only one reason, why i don't using Chrome. Shame for me, i know.

All you have to do is to download a file to %temp% without asking where to store it. 
Why does it take so long to implement such a useful function? 
I hate it to have to save a file continuously when I receive it via email, and 
afterwards I have to go to the download location to delete it. 
This is really annoying me!

To help myself a little bit I set the default download location to Desktop, so that I 
can quickly delete them. If I have a download I want to save I later copy it from 
desktop to it's destination-location. Because almost 75% of all downloads I make are 
only to view it once. Just 25% is for storage on my HDD. 
And because my %temp% location is automatically emptied every 2 weeks I wouldn't mind 
if these files are stored in there!

An other good example: all the screenshots attached to the bugs/requests on this 
website. Nobody does collect them (I think). But everybody has to save them to their 
HDD.

 Issue 9725  has been merged into this issue.

This is VERY annoying i unknowingly was forced to download an online installation 
package that was 232MB only to discvover 35 minutes later that it wouldnt work since 
it was looking for a local file on the site . SOLUTION: was to open IE and run the 
file from there no 35 min wait to download just a straight 10 min install.

its a shame, I Really like chrome but when simple Features like "print Selection" and 
"Open/Run from Location" are missing and are available in every other browser it 
makes me wonder if Google really wants a successful Browser, especially since these 
issues have been around since sept 2008 (Chromes Launch Date)

@sketch42, although this feature (ability to run files "directly") is important, it 
cannot help with the situation you described.  In fact, what you describe can only 
mean the website is doing something *different* for IE such that it works in IE.

The ability to run files "directly" still requires the file to be downloaded first.  
It just means Chrome should function differently eg. don't bother to show the file in 
download shelf but just automatically run/open the file after download.  If the file 
is 232MB in size and takes 35mins to download, it will take that long to download 
first before it can be "directly" run.  If the file requires a local file from the 
site, it will fail regardless of being run "directly" or after being saved first.  
(Most certainly, you are *not* running the program "from the site" regardless of 
saving first or running "directly".)

So if the program can run for IE and complete the install, the website is probably 
sending a different installation program that actually works when running in IE.

At the risk of going around in circles some more, doesn't this feature just require 
downloading the file to a temp directory, opening it as soon as it's downloaded and 
periodically cleaning up the old temp files?  I believe that's how other browsers do 
it.

@ghosttie, you are totally correct!  That's exactly what it needs and that's exactly 
how other browsers do it.

Which is why I point out to sketch42 that the time for a full download is still 
required and you are running the downloaded file from the temp directory (definitely 
*not* "from the site").

 Issue 10745  has been merged into this issue.

@krtulmay
However, the browser has full access to caches, and temporary file folders(and vice 
versa), and therefore can do what Sketch42 said.

@kidyelman, a browser's (including Chrome's) access to its disk cache and/or 
temporary folders can only perhaps help if downloading/"directly" running a file for 
the second time (assuming the downloaded file is in the cache).  It may be possible 
to "download" or "directly run" the file quicker if it is in the cache.  Otherwise, 
downloading a large file for the first time will always take all the time that is 
required to download the entire file.

Also, access to cache and/or temporary folders cannot help at all with the second 
part of sketch42's problem.  If the downloaded file requires a local file from the 
website to properly run, it will always fail regardless of downloaded first, or run 
"directly", or disk cache, or temporary folders.

I've just noticed that this issue is tagged as Milestone X and even DesignDocNeeded. 
Perhaps there's a misunderstanding... I don't know what you developers have in mind, 
but I think that simply doing what ghosttie suggested (comment 22) would do it. When 
I starred this issue, some time ago, I only meant this (and by reading the comments, 
I'd say I'm not the only one). "DesignDocNeeded" makes me think you mean something 
different and more complex. That's why I say there might be a misunderstanding.

And if you accept one piece of advice, I'd set it to Milestone 2.0: this is currently 
the only feature that I do REALLY miss, and though I'm not an expert coder, I think 
implementing this could be a lot easier than many other things.

I believe one issue is with files downloaded through secured channels (by that, I 
also mean using authentication) as you'd probably don't want your file to live in a 
temp folder until someone or something deletes it; it would be hidden to many eyes 
but recoverable by anyone.

And how about the "private" mode?

Let's say I'm using Chrome in an Internet cafe (yes, unlikely they'd use Chrome but 
imagine they'd do); I wouldn't want the files I download from my webmail to be 
recoverable by the next guy to use the PC. Storing the file in a temp folder isn't a 
solution (except for geek users). IE has an option (off by default!) that won't let 
it automatically store a file in a temp folder if it's been downloaded through HTTPS 
(you'd only have the "save" and "cancel" options, no "open" button in the dialog 
box), this is a first step, but still not enough (many webmails don't use HTTPS and, 
I repeat, the option defaults to "off", i.e. always save temp files).

 Issue 10593  has been merged into this issue.

 Issue 11707  has been merged into this issue.

 Issue 12994  has been merged into this issue.

http://docs.google.com/View?docID=ddwscr86_3gf6phhv6
Here is a good design doc for a new download/upload system, imo
Taken from  issue 4715 

(No comment was entered for this change.)

 Issue 18701  has been merged into this issue.

I did a poll in the Google Chrome's community most visited of Orkut.
After a significant number of users have voted, the count showed, for now, that the 
"Chrome way to download" is one of the features most bother the people.

This, of course, is making many users leave the Google Chrome. There's a thing 
important to do here.

I propose, then, some changes.

In the "Under the Hood" should be added one more option, like this:
(Options 1 and 2 already exist)

-------------------
Section "Downloads": 

Download location:
1. [C:\....  ] [Browse...]
  2. (  ) Ask where to save each file before downloading
  3. (  ) Ask for open file from a temporary folder

-------------------

To make life easier for lay users, who are accustomed to other browsers, the "2" and 
"3" should be marked by default, when Chrome is downloaded.

BELOW, I DESCRIBE MY PROPOSAL FOR BROWSER ACTIONS ON EACH OCCASION:

** a) 2 and 3 deselected: When the user clicks on a link to download, the browser will 
automatically save the file in the address selected in 1 (just as currently happens 
when the option is not marked 2). Chrome alert, as already do, if a file is dangerous.

** b) Only 2 selected: When the user clicks on a link to download, will be taken to a 
dialog box to select where he wants to save the file on his computer. After, the 
download is done normally.

** c) Only 3 selected: After clicking on a link to download, a dialog box will open:

------------------ 
You are trying to download the file 
"example.xxx" from "http://example.com" 

What would you like Chrome does? 
[ Open ] [ Save ] [ Cancel ] 

( ) Run automatically after download. 

Note: For details, change the downloads settings
at "Under the Hood" in the "Options" menu.

------------------

If the user clicks "Open", the file will be automatically saved on a temporary folder, 
and if selected the option to "run automatically", that opens automatically for now, 
just as if it were simply clicked on the downloads bar. 
If the user clicks "Save", the file will be downloaded normally, at chosen address in 
the 1.

** d) 2 and 3 selected: This should be the default. Then, when the user clicks on a 
link to download, will be taken to the same dialog box drawn above. 

If he clicks "Open", the file will be automatically saved on a temporary folder, as the 
topic "c". 
If he clicks "Save", will open another dialog box that asks where wants to save the 
file on his computer. 

Thus, we facilitate the adaptation of new users of the Google's browser. 

Note: Extensions can customize the browser action per-filetype to download.

A dialog box will appears only for those who are with both items selected (2 and 3), as 
shown above. 
So those who feel disagreeable, just need to clear the options, only once, in the 
settings of browser.

Obviously the problem here is that what is being proposed will add another step to the 
download process and Chromium's principle is to eliminate as many steps and prompts as 
possible. What I would suggest is.....

1. Add an option to the download shelve "Run and Discard" to delete the file after you 
run it.
2. Download the file by default into a temp folder and the add a "Run and Keep" option 
to the download shelve. 

"Obviously the problem here is that what is being proposed will add another step to 
the download process"

There would not be more steps if Chrome does this:

Note that Chrome already has an option to automatically open the file based on its 
extension. (After saving it in the default or chosen directory)

If the file isn't marked for auto-open, Chrome handles it as it does now.
Else, Chrome downloads the file to a temporary folder and opens it. (makes the 
default program used for handling that file type open the file)

So, no more steps involved.

What if you want to save the file after opening it?
The application used to open the file can usually save it. If it can't, add an option 
to the "right click menu" to save the file, that would really copy the file to the 
chosen folder.

When does Chrome delete the "opened" file from the temporary directory?
Just like the Firefox add-on "Download Status Bar", when you close the downloads bar.

Of course, as already mentioned, Chrome could also have an OPTION to show a pop-up 
asking to open or save.

(No comment was entered for this change.)

Download shelf needs improvement.  The suggested design doc seems great! And allows 
for Chrome to be a window manager also for documents that reside in the download 
shelf...

* The "Download Shelf" is where downloaded files go. The GC team wanted to allow you 
to upload files (think webmail attachment) directly from the shelf too; see this 
outdated page: http://dev.chromium.org/user-experience/downloads

* Currently the shelf can't come back if you hide it, and deleting files is too 
complicated.

* Sometimes you want downloaded files to be only temporarily on the current computer, 
but primarily on a storage service (S3, perhaps GDrive, ...). Thus you'd need the 
Shelf to integrate with the file storage service of your choice (S3 is one). The 
Shelf could become a "New files" folder. The Chrome page for Downloaded files could 
split (hor/vert) to show the file service on one half.

(And by now this issue has degenerated to a design doc discussion? Sorry for adding 
to the existing confusion...)

@simon.bohlin: This is not a thread about the design, but about being able to save the 
file to cache instead of to the "downloads" folder.

If you would like to discuss the design, I suggest continuing your discussion at  Issue 
23767  or Issue 9845.

 Issue 26783  has been merged into this issue.

 Issue 23668  has been merged into this issue.

I have set my download location to %temp%\ChromeDownloads\, so that they will 
automatically be saved into the temporary folder, and be deleted after a week. 
However, for large downloads (size > 256MB) I now use a self-created program that 
deletes the file immediately after you close the program.

But I don't understand: this issue is very old, and it is the 14th most requested 
feature. Why doesn't Chrome add it? It doesn't seem to difficult to me. In my own 
program I added this feature in about 3 hours, fully working, no files are left 
behind!

If they are afraid for security because the file stays in %temp%: They could create a 
process that deletes the files once they come free (no longer in use). They could 
also use the (annoying) googleupdate process for this. 
Furtherthemore: it is the users choise if he/she saves the file in a temporary 
location. 
Also: if it really is because of this security, then they should think: what is more 
dangerous? A file saved in %temp% or a file in My Documents? I guess they are both 
equal dangerous for private files.

 Issue 9023  has been merged into this issue.

This is a basic feature which should not be too hard to implement and is needed by a 
large amount of people. Saving all files..including unnecessary ones..then locating 
them..then deleting them..every time..can become very frustrating and annoying!

I agree this is a basic feature that is a must. It's very "un-google style" not to 
being able of having a quick look to a file (ie Open Document File) without manually 
saving it to a permanent HDD location.

How is this still not fixed? It's the only thing keeping me from using Chrome. This
feature request was first reported over a year ago and it's not a complicated
request. Frustrated.. :)

Search query for feature requests: 926 results.
https://code.google.com/p/chromium/issues/list?can=2&q=type:feature

[clearing CC list since stupid Google code is complaining about one of the usernames 
on it]

Search query for ABSOLUTELY IMPORTANT feature requests with over 150 stars:
https://code.google.com/p/chromium/issues/list?can=2&q=type:feature&sort=-stars

This issue is in the Top 10 feature requests!

How is this still not fixed?  This feature request was first reported over a year ago!
How complicated can such a crucial requirement be?

I agree. This is a priority function, it's not a minor option but a basic usability 
function. Imagine if you read online pdfs, download torrents, view videos (not 
streams), listen to audio streams through an external program etc. 

How many unnecessary files would exist inside the home folder each day?

Comment 58: "[clearing CC list since stupid Google code is complaining about one of 
the usernames on it]"
Does this mean nobody of the Chrome team will be notified about this and so this 
feature request will be ignored? I don't hope so.

Comment 60: "How many unnecessary files would exist inside the home folder each day?"
At least 10 to 12 per day in my case.

I absolutely don't like IE, but if you look at it's way how it downloads files it is 
much better (also the only point IE is better in)...

This full issue there has been ZERO explaining or other useful comments from 
@chromium.org users. All what's done is adding labels, CCing people and merging 
issues. What is the reason this very old request is still not fixed, as it is one of 
the most requested features? (Correct me if I'm wrong), but as far as I can see this 
request has been fully ignored by the Chrome developement team.

BTW: sorry for this useless post, but it is more useful than all of those 'issue X 
has been merged into this issue' emails I receive because I starred some issues. Also 
sorry if you think this post is too offensive.

(No comment was entered for this change.)

Adam, be sure to review the closely related  Issue 2292  when fixing this bug.  Thanks!

Fixed here on 4.0.249.30 / Linux ( you have to enable it in the options )

In another bug, nsylvain suggested that if the "Always open type" checkbox is checked, 
the download should go to a temp folder; else do what currently happens. This might 
be a good solution for this issue, as it doesn't add UI overhead and seems to solve the 
issue. The only problem is discoverability.

(cpu says "Be careful with that. If by opening you mean running, then in windows
an exe could end up using a dll that just happens to be next to it in
the temp directory.". I guess "Always open" works only with save files anyways?)

if we implement this, at the very least each download should have its own random folder 
inside TEMP. In windows programs tend to load dlls in the same directory over other 
directories, this would allow exploitation if you happen to have a malicious dll 
floating around in TEMP.

Or at least to keep things a little more organized, perhaps "each download should have 
its own random folder  inside a Chrome folder which is  inside TEMP".

But there is no question about "if".  This is a requirement and must be implemented.

Area-UI-Features label replaces Area-BrowserUI label

Oh and please allow to totally disable the download bar.

I currently use the "Always open files of this type" option for Internet Radio 
Playlist files (*.pls) which works great, it opens Winamp for the first 101 clicks.  
Shoutcast sends the same file name regardless of station (tunein-station.pls) so with 
Windows XP it can only rename so many files with same file name (ex. tunein-
station(100).pls)

This is a very annoying behavior that has not been fixed with Chrome yet, I never 
want these files stored in my Download folder as they are useless to me.  I moved my 
parents over to Chrome and they see this issue a lot and they don't understand why 
behavior changes over time.

This really needs to be fixed sooner rather than later.

Yep.  We need to figure out a UI for this feature.  I'll follow up with the UX team
after the holidays.

This issue has cause nearly half of the people I introduced to Chrome to switch back to 
Firefox.  This is getting absurd.  The status quo should be enabled, and security 
improvements or whatever should be made AFTERWARDS.  I have to manually pick through my 
downloads folder to eliminate useless files on a weekly basis.  
I know this comment does not add value, but the number of stars is NOT representative 
of demand.  Almost no one i know that wants this would have stumbled here and actually 
bothered to star or post.

Maybe add a (configurable?) list of file types wich should always be stored temporally 
(like .torrent, .m3u, etc.)?

Some files should not be stored at all. Instead an application must be invoked with url   
argument. It is necessary for example for streaming media or java webstart - launching 
local jnlp file usually works but is not 100% correct as it may miss codebase.

I don't really thing this is necessary simply because Chrome does downloads a bit 
differently already.  If a user wants to open a file, they can click the entry in the 
download try and it will open when it finishes, which is what they wanted.

I am finding it mildly annoying that so many people in this thread seem to think 
other browsers can magically open a file for viewing without having to save it 
somewhere on the drive first.  As others have mentioned, they usually go in %TEMP%. I 
dunno if other browsers clean up these files but there's little need; if disk space 
is low Windows comes through and clears up %TEMP% anyway.

Chrome's behavior is already very close to other browsers' open behavior but requires 
less prompts.  However to make it even closer without sacrificing the minimalist 
Chrome experience I have tried to come up with some ideas that may help:

1) Make the default download directory %TEMP%, or make a ratio button in options to 
use this instead of My Documents\Downloads.  Files desired for permanent download can 
be dragged off the tray by the user.  Files may be cleaned up periodically, however 
there are few reliable ways to be sure the user is "done" with a file, one way could 
be to check the system uptime on browser startup and delete all files older than that 
time.
2) As P.Redert suggested... Firefox has a list of mime types and associated actions 
you can configure.  Chrome could possibly use something like this, or at least 
extension APIs could be put in place to allow for extensions to control what happens 
with an individual file (handled with a plugin, downloaded, force display in browser, 
alter the mime type, block the file from being handled at all, etc).  This may be a 
bit too much for this bug but the APIs would be welcome by extension devs I think :).
3) Right click option on links right next to "Save link as...", something like "Open 
link in associated program".  Not sure if there should be a security warning for 
that, or where it would be appropriate to place one.  Download tray maybe, like for 
downloads.
4) Re-instate the "Delete downloaded file" option on downloads (I miss it!).
5) Download drop-down menu option to "Always remove old files of this type from 
download directory" or something similar, see #1 for my ideas on when files would be 
removed and the criteria for doing so.
6) Options dialog option for deleting downloaded files when download bar closed (also 
would happen on browser close).  A couple tweaks would have to be done IMO: Download 
bar close button could be made red to remind users it is now a destructive operation, 
undocking the last tab from a window with a download bar would move the download bar 
(or rather, the items in it) with the tab to whatever window its dropped in.

Some of these could also be applied per-mime type as in #2.

Now my responses about what others suggested:

@thakis: Same problem with downloading DLLs to My Documents\Downloads or wherever.  
Besides, if we're talking about a group that downloads strange DLLs from strange 
sites, we can only do so much for them in terms of security.  These are the people 
who turn off their malware filters because it blocks them from downloading a 
screensaver that nice spam mail linked them to.

@zibree: Nice idea but you should open a new bug for that to be sure it's noticed.

@artypan: Security is at the heart of Google Chrome and your suggestion goes against 
that core philosophy.  Google is not about the status-quo.  Furthermore this problem 
has nothing to do about security, but using the IE or Firefox UI would be exceedingly 
annoying since I am used to the Google UI.

@P.Redert: As a side note: uTorrent has a feature that can watch a directory for 
.torrent files, then load and delete them when found.  I've used it and am very happy 
with it.

@mwisnicki: That is handled by NPAPI browser plugins and is a separate system from 
downloads, it is not relevant.  All browsers will only download a file if they can't 
find a plugin that handles that type.

You wouldn't save every HTML file, YouTube Video, or image the user ever views, would 
you? Regardless of there being other (often hackish) solutions to this, the problem 
is the same. I went into my downloads directory today to try to find a file and there 
were literally thousands of torrents, documents, and PDF files in there that I had 
loaded once and never had any reason to ever want to load again. Often they were even 
multiple copies of the same document. It took me a good half hour to sort through 
them and delete the 95% that I had no intention of ever using again. That's just 3 
months' accumulation.

While in theory there are ways to rationalize not adding a feature to deal with 
temporary files, in reality not having it ends up making a big mess and leaving a lot 
of pointless work to the end user. Other browsers have all had to deal with and 
solved this problem, in some cases over a decade ago. Chrome's time will come as 
well. 

I wouldn't be too happy with many of the things proposed here. For example:

1. Choosing whether to open or save the file using the MIME type or extension: there are zip files I want to open and zip 
files I wan't to save to my computer. No, thanks, I don't want my browser to choose this for me.
2. Automatically removing *old* files: what? I don't want to have files randomly disappearing from my downloads folder just 
because they're old.
3. Automatically removing *new* files when closing the download bar or the browser. What?!
4. Removing downloaded files of certain types: see 1; I don't want to have two Downloads directories and to stay all day 
copying files from one to the other so that *my* browser won't delete them.

And so on. (though a way to delete them from disk using the downloads tab would be nice)

So yes, I *want* to be asked every time what I want to happen with my downloads. Anything else for me would be more or less 
equivalent with data corruption. I don't care about showing modal dialogs and I wouldn't mind clicking a button in the 
download tray to choose between "open" and "save".

@grayshade:
RE: 2 and 4. If you don't want it deleted, choose "save" instead of "open". The idea of 
"open" is that you do not want the file permanently saved on your PC.

The Open/Save feature is universal in virtually all browsers - I don't see why Chrome 
should be different. The lack of this feature is THE #1 reason that people I know have 
yet to switch. If you like, though, you can have a "settings" option to change the 
default behavior.

@#81
That's what I was insisting on, i. e. being asked every time.

Chrome is/will probably be different from other browsers because of certain design 
decisions regarding showing dialog boxes and asking the user about stuff.

@spet...: You hit the nail right on the head there. If I just want to open something, 
I don't much care where it's stored as long as I can view it and when I'm done it's 
not sitting somewhere on my computer taking up space. I would expect these files to be 
cleaned up at some point...in other words I don't want to have to think about them.  I 
expect overall that more files are downloaded by people just to be viewed than are 
downloaded for the purpose of retaining a permanent copy. If one wanted a permanent 
copy, they should be able to elect to explicitly save it somewhere so they'll know 
where the file is for later reference. 

 Issue 33582  has been merged into this issue.

@abarth@chromium.org & the Chromium team: Do you know if this feature is happening? I'm 
tired of checking this webpage constantly... and I guess a lot of people too. Please 
let us know if it is in Google plans to implement the open/save dialog and we should 
keep waiting or if we should stick to Firefox or IE... I'm not criticising, I would 
just like to know if you're taking all this people suggestions and time into account or 
if you're just wasting our time. I hope you take my comment the wrong way...

Sorry... I meant you don't take my comment the wrong way in my comment before.

Sorry for not updating the bug.  Here's the current status of this feature:

1) I've been talking with the UI team about how this feature should work from the
user's perspective.  We have a preliminary design that we're not that excited about,
but we have some hope that we can improve iterate and improve it.  Basically, the
design is to make "open without saving" the default download behavior, but give users
a "save" button on the download shelf that lets them keep the downloaded.  (Note that
we don't want to use the same modal user experience that other browsers use here
because one of the design goals of the product is to avoid modal prompts whenever
possible.)

2) While we're iterating on designing the user experience, I'm going to implement the
"back end," which is the code that actually knows how to write the download into the
temp directory and clean it up at the appropriate time.  That's on my list of things
to work on, but it has a lower priority than some other tasks in my queue.  You can
search for other bugs assigned to me and see what else I'm working on, if you're curious.

At a higher level, we understand that this is a top user request, and we're not
ignoring you.  However, without a complete user experience design, it's difficult to
target this feature to a particular Mstone.  As long as this features is targeted at
Mstone-X, I'm supposed to prioritize bugs targeted at nearer Mstones (like Mstone-5).
 I'm continuing to push for this feature, but I'd like to ask you all to have patience.

Thanks, and I'll try to keep the status of this bug more up-to-date in the future.

@88: I see the appeal to the "save to temp without prompting by default" idea, but 
that cold be a MAJOR pain for people with multi-drive setups who regularly download 
large files. No matter where you drop them, a non-trivial portion of the data would 
have to be moved to a different physical drive.

(I have three drives and I'm preparing to add a fourth)

You seem to be making an assumption about where the temp directory is stored.  One
possibility is to store it as a subdirectory of the downloads directory.

In any case, that's the easy part of the problem.  The UX is the real challenge here.

@Comment #89

Chrome could (and possibly will) do what every other browser does to handle your 
situation and allow you to download to specific locations by right-clicking links and 
choosing where to save to. Also, as mentioned in Comment 88, I think it also suggests 
the ability to define where to save a temporary file if you want to keep it 
permanently.

@89
Please correct me if I'm misunderstanding the current behavior, but when I click a 
download link with the "Always ask where to save" option enabled, it immediately 
begins downloading the file to *somewhere* even before I choose where to save. I 
witness this when I don't react to the download prompt for a long time and when I 
finally choose where to save, the file is complete already.

I presume the same behavior would be present with the open without saving scenario. 
The file would start downloading immediately when clicked, and then rather than 
presenting me with a "where do you want to save this" modal dialog, it'll just start 
saving to the temporary location like it currently does. I would then be given the 
opportunity to click some button on the download bar (while the file is still 
downloading, or after it is complete) to save it to a permanent location and the file 
would be moved to that location (if download is in progress, move existing data to 
final location and complete download to that location to eliminate the need to move 
the complete file).

@90: Actually, I'm just assuming my current behaviour ("Ask where to save each file 
before downloading").

Recognizing that, if the file must download completely before I can choose where to 
save it permanently, than no matter where the temp dir is, I'm going to end up 
waiting for roughly 50% of my downloads to be moved to a different physical drive if 
the browser doesn't give me the option to specify the destination before the download 
begins.

I'd rather not have to screw around with manually using wget for downloads (assuming 
the site allows it) just because it's quicker to open a terminal than to move a half 
my downloads because you can't simply relink inodes when performing cross-device 
moves.

I vote for this too. I want a general purpose "open with ..." added to the context 
menu. I dont want to have to download it, then seperately run a different app to open 
it. I want chrome to do that for me, so all I have to do is close the app's window, 
and resume where I left off (envision a page full of PDF links you need to review)

I dont want this to force mime-type either, even if the website borked it up and says 
app/octect-stream, *I* want to be the final say over what app opens the link.

And as long as I'm dreaming, I'd like to be able to choose between "run the app, and 
give it the uri of the link (so it can download it)", and "download the link, then 
run the app and give it the path to the file)

In fact, as long as I'm dreaming, being able to override a site specifying a embedded 
content (eg plugins), and just letting me link directly to the contact, to open a with 
NON-"plugin" application, would be nice too. The same for sites that want to check to 
see what plugins you have before letting you have some media - a local override would 
be nice. "No, I dont have Apples QT plugin - but pretend I do, and just give me the 
uri to the file and I'll use my non-Apple software to play it)

(No comment was entered for this change.)

(No comment was entered for this change.)

@andy16666 Not sure if you understand how the web works.  Every HTML page, youtube 
video, etc IS saved.  It's called the Cache.  They're just cleared out periodically, 
and it's nice and transparent to the user.  Some files skip the disk cache (large 
videos are not cached on disk) but they are still downloaded into memory.  In this 
case Chrome handles them internally with plugins and not external programs so no need 
to write them to disk.  Also I don't have problems with accumulation but I rarely 
download files I don't intend to keep, and I clean my downloads directory up almost 
every day.

@grayshade I was just throwing ideas out, I personally would not use most of them.  
Perhaps regex filters on the source url could be used for filtering downloads for 
more advanced users

@jeronimorv If you click the little star you'll get an e-mail every time a new post 
is added to this bug.

@abarth@chromium.org Might be best to just use the download directory itself, and 
just be careful about what you clean up when doing garbage collection on old files.\

@megadave That's sort of a separate issue, you're taking about Firefox's Applications 
tab in Options.  It would be nice for Chrome to have that sort of flexibility too.

You can't really "pretend" to have a plugin though since the way sites check for the 
plugin is usually to start an instance of the plugin and begin talking to it before 
opening any media.  In addition, the file handled by the plugin may not necessarily 
be the file you are interested in.  For example, flash players usually are just a 
simple SWF to play external media, specified in a parameter to the SWF.  Or maybe the 
parameter points to a playlist containing a list of urls.  Any other plugin can also 
be associated with a playlist file type with similar unpredictability.  It's really 
impossible to cover all bases.

Though AFAIK NPAPI does allow for the browser to tell the plugin "hey, here's the 
Cache folder, dump files to cache here", or maybe the browser plays a more active 
role in the downloading I dunno (never worked with NPAPI before).

@megazzt: I know very well how the web works. I'm a computer science graduate who's 
been working with web technologies for a decade and a half, and I've written hundreds 
of thousands of lines of code in dozens of languages consisting of everything from 
operating system kernels to corporate information systems. 

Over the years you learn the value of talking about the desired behaviors of software 
in high-level terms. In fact one of the stupidest things a programmer can do is to go 
straight to coding before they have a good idea of exactly what they want that code 
to do. It generally leads to an unmaintainable mess, which unfortunately explains a 
large portion of the software out there. 

In this case I think the Chrome developers should be the ones making the choices 
relating to how this feature is best implemented at the low level, since they are the 
ones most familiar with the particulars of how Chrome works. Also, personally just 
because I'm a computing professional doesn't mean that I want to deal with the 
particulars of low level technologies when I'm doing regular stuff surfing the web, 
downloading torrents, or viewing PDF files. 

The ideal behavior here would be for files which are opened by default to be cached 
temporarily, for exactly the same reason images and you tube videos are cached 
temporarily. The user can click "Save As" from their application or right click on 
the link and "Save As" if they want the file to be preserved in a particular location 
on disk. 

Ok,

What seems to be the issue here is that certain people do not seem to realize that various 
users have differing wants and needs when it comes to a browser.

1. AUTOMATICALLY OPENING FILES: While this is a nice *default* option, there is no sense in 
forcing this upon all users. For the users that would like to be prompted whether [or where] 
to save each file, there should be an option for this. Yes, users can right click and "save 
file as...",  but this is not always practical. Many times a user thinks that they are 
accessing a web page, when in fact the link points to a PDF or DOC file. Without the option to 
be prompted, the browser would automatically open the file when the user clicks it, forcing 
the user to close the program and re-download the file via "right click". Alternatively, 
document links on many sites link to a page with a JavaScript or meta redirect to the 
document. Right-clicking these links and "saving as" would just download the html page instead 
of the document. While these may not be an issue to some, it is to others. A quality 
application takes all user needs into account.

2. TEMPORARY DOWNLOADS: Yes, we all understand that *all* files are downloaded to the 
computer. However, when using the term "open" instead of "download", people refer to "temp 
folder downloads" and "permanent downloads". While *you personally* may not have such issues, 
keep in mind that many people download some files for one-time use and others to access later. 
Additionally, there are many users who do not understand what happens *behind the scenes* and 
just want a working web environment. Yes, ideally all users should be aware of how these 
things work, but they don't and they won't. This is not a "geek" browser. This is an 
application for all users.

3. MODAL DIALOG ISSUES: Yes, the elimination of modal dialogs is a nice feature. But what harm 
is putting alternative options in the "settings"? Also, Chrome has done a nice job with 
eliminating the modality in the login dialog box (i.e. you can still switch tabs while the 
login prompt is displayed), why not apply that to the download dialog(s).


Bottom line, just because you don't have a personal desire for a specific feature, does not 
meant that it is a bad idea. Chrome is a great browser - faster than any other that I use (and 
being a web developer, I use all major browsers). Why deprive it of features that are so 
clearly bothering a significant amount of users?

 Issue 36389  has been merged into this issue.

abarth@chromium.org:

PLEASE do not implement a system like that (files automatically being run). Nearly all 
the files I download I don't want to run immediately.

@theodorejb

We want a system that let us choose if we want to open or save the file, just like IE.

abarth@chromium.org:

Please DO implement a system to run files automatically.  Nearly all the files I 
download I DO want to run immediately.

What is going to be implemented is an OPTION to RUN/OPEN files that you do not want to 
save.
NOT automatically opening/running or saving all files.

This is the feature I use (well used) the most..Please fix this..it has been a over a 
year now and all other browsers have it..this feature and the master password feature 
is the only reason I keep Firefox sadly.

Please give an option to simply "open" files. I'm using chrome on Fedora 12 and I 
usually open a lot of PDF's to read, but I only want to save those which are relevant 
to my work. Incredibly chrome enforces me to manually delete every PDF I don't want, 
and that is REALLY ANNOYING. The way Firefox handles saving/opening files is a lot 
better.

 Issue 36975  has been merged into this issue.

There is a very simple way of fixing this.

All that is required is an 'Open' button next to the save button and when the 'Open'  
button is clicked after a download is initiated, the file is downloaded to the %temp% 
directory and automatically opened when completed.

The save button can still do exactly what it has always done and the options in 
'Under the hood/bonnet' can remain exactly the same.

That is all that it needs to do, that cannot be hard to implement surely?

If for some reason the "open" action is not wanted and the "download" action is the 
only option by design, would be nice at least to be able to move or delete the 
downloaded files quickly (probably from the download bar). This basically would allow 
to do somewhat the opposite to what happens on other browsers but with the same 
result:

Other browsers: User opens a file, application X starts opening the temporary copy.
- User is not interested in keeping the file (closes the application).
- User is interested in keeping the file (save as: from the application, he can 
define name and location of the saved file).

Chrome: User downloads a file and clicks on the button in the download bar, 
application X starts opening the saved copy.
- User is not interested in keeping the file (closes the application, from the 
download bar he can quickly deletes the file)
- User is interested in keeping the file (has to do nothing unless he wants to move 
the saved file somewhere else, from the download bar can quickly move it).

It's a different approach, I am not sure if in practice it would work or would be 
better than the other browser's way.


PS: maybe I am missing something but the revision mentioned in comment 117 doesn't 
seem to be related to this issue.

(No comment was entered for this change.)

This has been going since Sept.2008. When will this be implemented? Its such a simple 
thing and SO many people want it... How many people commenting does it take for Google 
to just do it?

This bug is targeted at Mstone-6.  If you're watching our release cycle, you might
estimate that Mstone-6 is roughly eight months away.

Why is it changed from milestone 5 to 6? How many people do you need before you add 
this 5 minute code feature? Few hundred people who starred this issue isn't enough? Do 
we need a petition to get Google to spend 5 minutes fixing this? Ridiculous

Come on guys. Be patient. I know it's the most requested feature but I think it's 
always been an issue of ui design, not backend code. You can appreciate that such a 
feature COULD change the fundamentals of how Chrome works (one-click downloads) and 
can't be implemented without due consideration. Besides, why should they hold up M5 
waiting on one feature? 

Why can't this be done through right-clicks, at least temporarily? I.e., 

 * Left click = one click download (as currently done)

 * Right click = Pop up menu with "Open" action that downloads the file to the temp 
dir rather than the download dir.

That sounds so simple this could even be done with an extension...

I can't belive how people loses that much time saying the same thing:
1) How this feature is SO important that they can't live without it and are moving to 
other browsers because of it. (I just find it a bluff)

2) How people say "such a simple thing" or "such a stupid feature".... 
Have you ever even programmed a calculator? Have you got any remote idea of what does 
programming this feature mean? Probably not, so please quit the whinning...

Instead of saying this stupid things, why don't you say something useful?

I doubt google reads this post, but I'd recommend them to do the following tweak 
until the feature is fully developed and tested for every OS:

Since when im currently downloading a file, I have the option to Cancel it, and it 
deletes the unfinished file, I assume that the ability to delete files isn't hard to 
implement. So, why don't we put an icon or a contextual item in the arrow that gives 
the user the option to delete the downloaded file? That way I'd not mind if it's 
saved in my Downloads folder instead of a Temp folder, because I can easily and 
effortlesly delete the file, without having to go all the way to the file's folder 
and manually deleting it.

Just my 2 cents

abarth@ - When fixing this issue (after 5.0 but hopefully BEFORE 6.0 as I have been getting more complaints from my 
mother about this issue recently..lol), please do not run all downloads automatically. All we want is something 
simple, the option to RUN/OPEN or SAVE. Thanks in advance.

Is it not possible to get this done with an extension?

(No comment was entered for this change.)

@ddeamilivia

1)Yes, this feature is so important that we can't live without it. I almost switched 
browsers as well because of this, but I didn't because I'm too lazy to change. 

2)Programming a calculator is NOT hard. Unless you use a language that does not have 
math functions, which I don't think exists, you can program a basic calculator in 
less than 100 lines. 

I'm not sure if YOU understand how easy this feature is. 99% of the work is already 
done. The 1% work is to make a dialog box asking if you'd like to open or save, and 
just delete the temp file afterwards. It is certainly MUCH MUCH MUCH MUCH less hard 
than implementing the language detection feature that Chrome just got, and this is 
100 times more useful.

Because I'm still a student learning computer programming, if I am a programmer, I'd 
GLADLY spend a day adding this feature in. I'm not saying this program takes a day, 
it just takes a day to read the code and see where to implement it. The actual 
programming cannot be hard. I see this as a homework, even a test level difficulty 
question. 

Currently, when I click on a PDF file, it *does* "run" and open the file 
automatically.  It also saves it in my Downloads directory and does *not* show up on 
my download bar.  Basically, this is exactly what I want except the file should be 
saved to my tmp directory.

Installed today for reading HTML distance ed materials (on a CD, many links to word doc 
readings as well as URLs.). Lack of such a function has turned what was a trivial task 
in firefox/IE into an awkward chore.

Can anybody just make this an extension for the time being?  I'm not sure if that's 
possible.  One alternate solution would be to make it work like .exes do, and have a 
prompt right on the download bar for open or save.  Open would save to a temp directory 
and save would save to your default location.  You could also have a checkbox saying 
something like 'don't ask me again,' 'do this action every time,' or something for each 
different type of file.

I can't believe the confusion this request has caused over the months. This is a
feature that every other browser I'm aware of has from its very first versions. A lot
of folks seem very confused even about what's being asked for, as though none of them
have even used another browser before.

What's being asked for ? It's not rocket science - Firefox handles it perfectly.
Executable files are a special case in Firefox - for security reasons, it only gives
the option to save the file. For file types known to be associated with an
application, the browser should give a simple Open/Save/Cancel dialog. Open means
save the file into temp files, then directly open it with the requested application.
Save behaves exactly the same as it does now - you can even keep the 'run on
completion' option if you want. Cancel is self-explanatory I hope.

Is it needed ? Imagaine this scenario. You try out a new word-processing package.
Everything seems to work great. Then you come to print your document, and find that
you're expected to 'save' the printout to a file somewhere, then open that file in a
separate print routine to print it out to a real printer. Oh, and then you have to
remember to delete the file you just saved, as it doesn't do that for you either.
Does that sound like a throwback to the 80s anyone ? Now you know why this particular
'feature' of chrome is so annoying to so many people.

This has gone on too long. I think the best thing to do is keep the default behavior 
and add an option to "Open link" in the context menu. Those who want to open instead of 
save would right click and choose open and those who want to download would continue as 
normal. It's no longer a method than clicking and then having to choose what to do with 
the file but at least this way it won't interrupt all user.

(No comment was entered for this change.)

This is going into Mstone-X until we have a UI design that satisfies our other Chrome constraints (not prompting, etc), and it's not something that is currently on our list. We do understand it's a commonly requested feature however so it is possible we'll consider it in the next 6-12 months.

 Issue 39773  has been merged into this issue.

(No comment was entered for this change.)

 Issue 58663  has been merged into this issue.

 Issue 68460  has been merged into this issue.

(No comment was entered for this change.)

 Issue 99284  has been merged into this issue.

 Issue 136178  has been merged into this issue.

(No comment was entered for this change.)

 Issue 145261  has been merged into this issue.

(No comment was entered for this change.)

Wow, Started:baueb is very out of date.  Moving back to available.

(No comment was entered for this change.)

(No comment was entered for this change.)

(No comment was entered for this change.)

 Issue 175713  has been merged into this issue.

 Issue 224022  has been merged into this issue.

 Issue 442149  has been merged into this issue.