## sa_figure2

## Source details

**Canonical URL:** [sa_figure2](https://www.imf.org/en/publications/gfsr/issues/2016/12/31/sa/sa_figure2.csv)

## Other formats

- [Markdown version](/en/publications/gfsr/issues/2016/12/31/sa/sa_figure2.csv.md)
- [Structured JSON version](/en/publications/gfsr/issues/2016/12/31/sa/sa_figure2.csv.json)

---

### Dataset overview
- Worksheet name: "csv"
- rowCount: 439
- columnCount: 1
- File type as supplied: dataset (XLSX or CSV)
- Canonical source URL provided: https://www.imf.org/en/publications/gfsr/issues/2016/12/31/sa/sa_figure2.csv

### Content findings
- The single-column worksheet contains HTML content for a web error/404 page rather than structured numeric or tabular dataset rows.
- Embedded HTML metadata and scripts are present, including doctype and <html> head/body content.
- Multiple meta tags and hidden input values are present with exact values observed:
  - meta name=robots content=noindex, nofollow
  - meta name=parsely-pub-date content=2025-08-26
  - meta name=Date content=2026-08-28
  - hidden input id=imfdate value=2025/08/26@10:31:22+00:00/
  - hidden input id=imftype value=Standard Page Full Width/
  - hidden input id=imflanguage value=ENG/
- The worksheet includes JavaScript fragments and analytics calls, for example:
  - "function gtag() {"
  - "dataLayer.push(arguments);"
  - "gtag('js' new Date());"
  - gtag('config' 'G-CPJPTDPL31' { 'page_location': decodeURI(window.location.href.toLowerCase()) });
- The sampled rows show site navigation and social links markup, for example:
  - "Follow The IMF on X"
  - "Follow IMF Facebook"
  - "Follow IMF LinkedIn"
  - "Follow IMF Instagram"

### Technical assessment
- The CSV/XLSX contains non-data HTML content (a 404/error page) — unsuitable for direct numeric analysis or ingestion as a GFSR figure dataset.
- Key indicators that the file is an error page:
  - Presence of "<title>404</title>"
  - meta property=og:url content=https://www.imf.org/en/errors/404
  - Canonical link in the HTML: https://www.imf.org/en/errors/404
- The worksheet sample rows include long HTML/JavaScript strings rather than comma-separated numeric fields.

### Recommended actions for users and pipelines
- Do not treat this file as a valid numeric dataset until the source issue is resolved.
- Attempt re-download or request the correct CSV/XLSX from the canonical publication location.
- If automated ingestion is used, add validation to detect HTML/404 content by checking for tokens such as "<title>404</title>", "meta name=robots content=noindex, nofollow", or presence of "<html" in the first column, and flag the file as invalid.
- Preserve original file and metadata (rowCount: 439, columnCount: 1) for auditing while blocking downstream numeric processing.

*Source: sa_figure2 dataset (worksheet "csv") — extracted HTML/404 page content as supplied.*

---


_Source: https://www.imf.org/en/publications/gfsr/issues/2016/12/31/sa/sa_figure2.csv_