Each format conversion has different rules and edge cases. The 5 most important things to know:
@key; repeated sibling tags become JSON arrays.address.city).colspan do not.Data lives in many formats simultaneously. Your database exports CSV, your legacy API returns XML, your frontend expects JSON, and your documentation platform needs Markdown. Moving data between these formats without errors requires understanding each format's rules and where they differ.
CSV (Comma-Separated Values) is deceptively simple, but real-world CSV files come with complications that trip up naive parsers.
Header detection: Most CSV files have a header row. Our CSV to JSON Converter treats the first row as keys for the resulting JSON objects. If your file lacks headers, add them before converting.
Type inference: CSV has no type system — every value is a string. A good converter infers types: 42 becomes a JSON number, true becomes a boolean. Our converter applies smart type inference while letting you override when needed.
Quoting and delimiters: Fields containing commas must be quoted. Fields containing quotes must escape them by doubling. RFC 4180 defines these rules. Tab-separated (TSV), semicolon-separated, and pipe-separated variants also exist — our converter auto-detects the delimiter.
XML and JSON represent the same concepts differently. Understanding the mapping rules prevents data loss during conversion.
Attributes vs child elements: XML attributes have no direct JSON equivalent. The standard approach maps them to keys prefixed with @: {"@id": "42"}. Our XML to JSON Converter follows this convention.
Array detection: Multiple sibling elements with the same tag name are implicitly an array. Our converter detects this automatically.
Namespaces: XML namespaces are common in SOAP APIs and enterprise XML. The prefix becomes part of the key name in JSON.
Text content with attributes: An XML element with both text content and attributes maps to a special #text key for the text content.
Converting JSON to CSV requires making a two-dimensional table out of potentially multi-dimensional data.
Simple arrays: A JSON array of flat objects converts directly to CSV. Each object becomes a row, each unique key becomes a column.
Nested objects: {"address": {"city": "London"}} must be flattened. The standard approach uses dot notation: address.city becomes a column header. Our JSON to CSV Converter handles this with the Flatten nested option.
Arrays within objects: {"tags": ["js", "python"]} has no clean CSV representation. Options include joining values with a separator or using a separate CSV for the array.
Converting HTML to Markdown is common when migrating content between platforms — from a CMS to a static site generator, from a web scrape to a documentation tool.
What converts cleanly: Headings, paragraphs, bold/italic, inline code, code blocks, blockquotes, lists, links, and images all have direct Markdown equivalents.
What loses fidelity: Custom CSS classes, complex table layouts, nested lists deeper than two levels, and any semantic HTML that Markdown has no syntax for.
Tables: Standard HTML tables convert to GitHub Flavored Markdown (GFM) table syntax. Complex tables with merged cells cannot be represented in Markdown. Our HTML to Markdown Converter handles standard tables and normalises whitespace output.
CSV→JSON コンバーターを使ってください。ファイルはブラウザー内のクライアント側で処理され、サーバーへのアップロードサイズ制限はありません。約50 MB を超えるファイルには、jq や Python の csv モジュールといったコマンドラインツールのほうが効率的です。
最も多い原因は、XML 属性が削除されることです。コンバーターで属性が保持されるようにしてください。また、XML 名前空間も確認してください。ns:item のような要素は、JSON ではそのままのキー名になります。
1つのツールで直接は変換できません。まず CSV→JSON コンバーターで CSV を JSON に変換し、次に JSON→XML コンバーターで JSON を XML に変換してください。2ステップですが、同じブラウザー内で完結し、データはアップロードされません。
Markdown は意図的に機能が限定されており、一般的な書式ニーズの80%をカバーします。CSS のスタイル、カスタムクラス、複雑なレイアウトには、Markdown に相当するものがありません。
REST API の標準は JSON です。パーサーのサポートが最も充実しており、一般的な構造化データではペイロードサイズが最小で、JavaScript でネイティブにサポートされています。XML は SOAP API やレガシーな企業システムとの連携でのみ使ってください。