I have my own directory for static files and that's why I want to move my favicon, robots.txt, sitemap and other files of the same sort to it. Therefore, these files will be accessible not as my_website.com/robots.txt, my_website.com/favicon.ico, but my_website.com/my_files_path/robots.txt, my_website.com/my_files_path/favicon.ico
I wonder, won't this hurt my website in terms of SEO? Are there other drawbacks? Is it better to put these files in the root directory?
Web robots will look for the robots.txt in your root directory. So it's a good idea to place it there. On this page Google says:
In order to make a robots.txt file, you need access to the root of your domain.
Most browsers will look for a favicon.ico in your root aswell. You can specify a different path to your icon with the <link> element. However, remember that all files on your site is not HTML. If a user views an image, PDF or similar then the <link> element will not work. In such cases, the browser will likely fall back on the file in the root folder.
Related
Please bear with me since I am a noob at html.
Let's say I have a local directory called website saved inside my local Downloads folder, and inside this website directory I have an html file called page.html
Inside the website directory, I also have another directory called folder
Inside the folder directory, I have an html file called page2.html
In the html code for page.html, there is this line of code:
page 2
When I open page.html locally in a web browser, the file path name is file:///Users/myuser/Downloads/website/page.html.
When I then click that page 2 link on the webpage, it brings me to file:///website/folder/page2.html instead of the correct path file:///Users/myuser/Downloads/website/folder/page2.html so it doesn't work.
I know I could change the href link in page.html to file:///Users/myuser/Downloads/website/folder/page2.html but I want this link to work even if I move the website directory into a different local directory. For example, the href link would work whether I have the website directory inside my Downloads, Desktop, or Documents folder, or even if I saved this website directory onto a different PC.
Is there a way to word the href link so that this can happen?
You’ll need to use href="../page.html" in your page2.html file.
I recommend you read up on what a URL is, especially the part Absolute URLs vs relative URLs
So a URL has different parts, beginning with the scheme like file or https.
Image from Mozilla at the before-mentioned URLs
You can skip certain parts at the beginning of a URL, which will give you a relative URL. These parts will be replaced by the user agent (the browser) from the current location.
For example, you can use scheme-relative URLs:
<a href="//myhost.example.com/page.html">
If the page containing that code is served via https, the link will also be completed to https: https://myhost.example.com/page.html. If it’s served via ftp, it will complete to ftp://myhost.example.com/page.html.
The same goes for other parts, and when referencing other pages from the same site, you would use path-relative URLs.
Absolute and relative paths
Now, concerning the path part of a URL, there is also a distinction between absolute paths and relative paths, just like in your operating system.
<a href="/page.html"> is using an absolute path. This means go to the root directory of the same drive or host, to find page.html.
In your case, the page2.html is delivered from file:///Users/myuser/Downloads/website/folder/page2.html.
So when you use a path beginning with / (absolute), it will refer to the root of the drive, so basically C:\ and complete to file:///page.html.
<a href="../page.html"> now is a relative path, relative to the current location. It’s saying go up one directory to find page.html.
So with the same location as before, this will complete to
file:///Users/myuser/Downloads/website/page.html.
I have read on several websites that using the link element is not beneficial/necessary if the favicon is already stored in the root directory of the domain under the name "favicon.ico".
However, if the file suffix, name or directory path differs, the link elements will be crucial.
Did I understand it right?
Exactly. The favicon in root and no link is the approach used by Google itself. But there are cases where you cannot place the icon in the root directory or you don't want to pollute this special directory. In that case, a link markup is necessary.
What is the correct way to do this and have what is generated / used by the browser be relative paths?
I need to do this in order to generate PDF's with links that will open files in a folder directory on a user's hard drive. The links must use relative paths, since the files will be downloaded in a directory structure that will be moved around on the hard drive. To further clarify the use case, the PDF(s) and the files will be downloaded in zip archives together, so the initial relative location will always be known.
I understand there are things in place in browsers to prevent this working from within a browser window, but in this case it will only be used in anchors in PDF files generated from HTML.
One of the problems with using something like this is that the protocol and domain is appended to the front of the path in the anchor. The relative path that's needed is to navigate within a folder directory in Windows Expolorer.
file.txt
The PDF's are converted from HTML using SelectPdf, and this is an Asp.Net MVC web app.
I have tried variations of this, which windows explorer is not able to use:
file.txt
All,
I did a bit of research but haven't found an exact thread or resolution to this issue.
I am using express in this webapp, Chrome Version 60.0.3112.113, and Win 10 Version 1703.
I am currently developing a site where I want to use a hamburger svg for mobile navigation. This is how the html sits for the "topbar"
<div id="topbar">
<img src="../images/hamburger.svg" alt="ham">
</div>
And here is the file structure:
https://puu.sh/xxDih/c842297b54.png
According to the structure, I should only need to do ../images/hamburger.svg, but when I do that, it comes up with a 404 error in the waterfall. I have run into this issue multiple times doing any sort of HTML sourcing into parent directories, but in JS files it works fine.
I'm not exactly sure what the issue is.
For the express server，every uri are processed by the express contains resource url and request url.
Request url(api) is refered to your express api config
resource(image, js, css, html...) is relative to your static server's root directory which was defined by using express.static(root_path).
That's what I want to say.
I noticed that images folder, node_modules folder, and pages folder are all in the same directory, and css is under the pages folder.
"../images/hamburger.svg" is the correct relative path from the pages folder, but being (big red flag) the node_modules is at "../node_modules/" I'm thinking that the server is serving from /pages/ folder, the servers root directory. meaning anything above the /pages/ folder will not be shared.
Clearly you do not want to share out ../../../windows/system32/ or the user documents etc. To prevent sharing those the highest directory you can access from the html page through a browser is the server folder being used. I'm thinking /pages/home.html is localhost/home.html and localhost/css/ is your css directory.
Programs running on the server can access files above the served directory, but the browser can not. "/node_modules/" should be outside of the servers root directory.
I realized this is an issue with express itself.
If (in this case) you have your index.html as express.static('./pages'), then it can't see anything above pages and considers pages as the working directory.
Me, coming from React (which stupidly was the first thing I learned even before basic JS), wants to put all the pages in one folder, which I think would make sense.
The workaround I did, which may not be optimal, was by putting index.html as a sibling to pages, css, and images in the src folder. Then in index.html, it has a meta tag as follows: <meta http-equiv="refresh" content="0; url=./pages/home.html" /> to redirect to the home.html page.
Again, this may not be optimal, but for a kinda OCD guy like myself this makes sense.
Update:
What we ended up doing is to have index.html be a static page, and then load the individual pages in an iframe. This website is mainly for information and has no database (yet), so there won't be much to process. Here's the new file structure that works.
http://puu.sh/xy5Dw/4dbc72ec06.png
src is now our working directory (express.static('./src')) and everything is detailed within there.
Once we do include a database, it will at most be 10 values in the server and will be using very basic requests, nothing crasy.
The physical folder structure on my web server is:
inetpub
oo|
oo|_wwwroot
oooooo|
oooooo|_MyApp ----> this is the root folder of my web application
oooooooooo|
oooooooooo|_images
oooooooooo|
oooooooooo|_styles
oooooooooo|
oooooooooo|_pages
I have an html file (test.html) under the pages folder. I want to link to other files (eg. stylesheets and images) via absolute paths. So, I have a line of code similar to this:
<img src="/images/roundbutton.png" />
When the page is rendered, the image is not displayed, as the "/" tells it to look for the image in a folder wwwroot/images.
Is there another way to use absolute paths to refer to the application root, instead of the site root? Or is using relative paths the only other option?
Yes, use the <base> element.