Scan again and publish by itself
Your site changes. These settings keep your translations up to date without you doing the same steps again.
Scan my site now: finding your pages#
On the site page, Scan my site now reads the pages in the box. Also find pages is ticked by default. Then the scan:
- reads your robots.txt and your sitemap (sitemap.xml, a sitemap index, and compressed sitemaps are fine);
- scans the pages you listed, then the pages in the sitemap;
- follows the links between your own pages, a few clicks deep, until it reaches Most pages to scan (25 to 500, 100 by default).
It never leaves the addresses you allowed for the site, never reaches private networks, never opens a page your robots.txt blocks, and waits a moment between pages so your site is not slowed down (if your robots.txt asks for a longer wait, it waits that long). Untick Also find pages to scan only the pages in the box.
While it runs you see how many pages are done, how many it found, how many new strings and how many pages it skipped. When it is done, the line addresses were skipped. Why? opens a list that explains each group in words: blocked by your robots.txt, on another website, not a web page, search or filter links (with a question mark in the address), too many clicks away, or left for the next scan because the page limit, the time limit or your plan's string allowance was reached.
Save page list keeps the addresses in the box without scanning now. The scheduled scan uses that list as its start pages.
The Automation box#
On the site page, the Automation box has two settings. Both are off until you switch them on, and each site has its own.
- Publish automatically after each translation run. When a translation run finishes, the site is published. A publish where nothing changed is skipped. Your plan's limits still apply. If the plan does not allow a publish, you see the failure in the notification list.
- Scan my site again. Choose every day or every week. The product scans the pages from your last scan and picks up new text. It does not translate that text for you by itself, unless you start a translation run. Choose Only when I press the button to turn it off.
A scheduled scan is the same scan as Scan my site now, and it also looks for new pages (see below). It follows your robots.txt file, only reads the addresses you allowed for the site, never reaches private networks and counts towards your plan's limit for strings. When the limit is already reached, the scan fails and the notification list says so. When it is reached part-way through a scan, the scan stops there, keeps the text it found and lists the rest as skipped.
Unpublished changes#
When translations changed after the last publish, the site page shows a banner with a Publish now button. The banner is exact whatever the size of the site. See Publish changes.
Notifications#
Notifications in the menu lists what happened in the last two weeks:
- a translation run finished, with the number of strings translated;
- a publish finished;
- a scan finished, with pages scanned and new strings found;
- a job waiting for translation capacity. It carries on by itself and nothing is lost;
- a job that did not finish;
- a payment that did not go through. Your published pages stay online;
- your translation allowance used up, and the date it comes back.
A number next to the link shows what is new in this browser. Opening the list marks everything as seen in this browser only. The list is built from your sites' activity, so it never disagrees with the pages it links to. It does not send e-mail. Reviewers' requests and comments are not in this list.