AnqiCMS (AnqiCMS) Template Tag Manual
Automate workflow through setting the scheduled release time of content, with draft links carrying the parameter preview=true for internal preview. Deleted documents can be recovered by moving them to the recycle bin. Understand the flow rules of four states: formal documents, drafts, pending release, and recycle bin to improve team collaboration efficiency.
Automatically obtain information from external websites through the content collection module, combine batch import functions of ZIP compressed packages and Excel tables, utilize the duplicate checking mechanism of the /api/import/archive interface to avoid repetition, and gradually build a structured corporate knowledge base.
Ensure consistency across language versions through full-page HTML translation,配合hreflang标签avoid duplicate content issues, configure independent keyword libraries for each language, and achieve true SEO localization rather than simple machine translation.
After changing the server IP or connecting to CDN, the review sequence is a three-level progression: first confirm DNS resolution (prematurely reduce TTL, check the effective range), then verify the certificate chain (check both sides of the edge and origin, align the origin protocol), and finally converge the collection side - the 301 redirect mapping cannot be interrupted, actively push the core pages to Baidu and Bing, and check the source address in the log.
When the content is not changed but the page loads slowly, the order of checking the cache control header is to first see which layer the instruction acts on: max-age controls the freshness seconds of the browser's private cache, s-maxage controls the shared cache of CDN and will override max-age, no-store makes any cache not store. This article gives the binary method of transport layer and rendering layer, as well as the checking actions of each layer: browser, shared cache, and origin server.
The site-wide replacement can change keywords and links in the main text in one go, but it only covers the content text layer: column names and address forms belong to the structure and routing layer. After the change, the keywords in the anchor text library and the matching paths of the static rules may not match the new address. At the end, three things need to be done - check the anchor text distribution range, review the static rules, and check the jump between new and old addresses to prevent broken links left by batch changes. Back up before changing, and keep a fallback ready.
The content has changed, but the front-end is still showing the old page. Most of the time, it's not that the program hasn't taken effect, but rather that the intermediate cache layer is still hitting the old copy. Investigate in three steps: first, prove false on the source station and confirm that the change has really been saved to the database; then clear the cache at an appropriate granularity, with edge nodes supporting a variety of ranges such as address, domain name, tag, prefix, and the entire site; finally, check whether the cache status header in the response has changed from a hit to a source, and confirm that there is no trace of verification through the user side. Make a backup before making batch changes and keep a fallback.
The two basic instructions of robots.txt are Allow and Disallow, which respectively indicate allowing or prohibiting the crawling of certain paths; the asterisk is a wildcard that can match any character sequence, and the dollar sign at the end of the line indicates that the address ends exactly at this point. This format is standardized by RFC 9309. It should be noted that it only constrains crawling behavior, does not determine whether a page can be included, and is not effective for all robots; after changing the pseudo-static rules within the site, the robots configuration needs to be reviewed to ensure that the match is still valid.
The standard text for excluding by robots, the fields only include User-agent and Disallow groups, nofollow is not defined in robots.txt, and it will not be parsed according to the standard. This article distinguishes between blocking crawling and blocking inclusion: crawling is controlled by robots configuration, inclusion is based on page-level control and server-side redirect convergence, and on-site robots settings and 301 redirect each have their own responsibilities.
According to Baidu Search Resource Platform documentation, when the API push is successful, it returns the four fields: success, remain, not_same_site, and not_valid: remain is the remaining quota for the day, and not_valid is the list of invalid links; when it fails, it returns client error codes starting with 4 or a 500 server error code. This article discusses the troubleshooting order by field and explains which part of the link push and sitemap covers the link inclusion path.
