Far HTML 3.3.1 serial key or number

This page is about converting a fraction (i.e. a ratio of two numbers, also called a rational number)into a decimal fraction and the patterns that occur in such a decimal fraction. It is interactive and you can usethe calculators on this page to investigate fractions for yourself to many decimal places. No special knowledgebeyond decimals and division is required.Some but not all of the Calculators on this page are combined into a single Calculator on Fractions - Decimal Calculator .

---

Changing a Fraction into a Decimal number

Converting a fraction to a decimal is just a division operation. Sothe fraction ¹/₂means 1÷2.
When we do the long division (or use a calculator!) we find 1÷2=0·manicapital.com fraction was easy - just one digit and we are done.

Some examples and patterns

People have always been fascinated by the patterns in decimal fractions, trying to find the order in the seeming chaos and intrigued by the sequences.

When we make a table of the first few reciprocals of the numbers 2,3,, that is when we turn the whole numbers upside down: from2=2/1, 3=3/1, 4=4/1,to 1/2, 1/3, 1/4, we get the following:

Three types of decimal fractions

1. In the table we can see that some decimal fractions stop after a few decimal places - those in the left-hand columns - such as 1/2, 1/4, 1/5, 1/8.
   These are called terminating decimals.
2. But others become an endlessly repeating cycle of the same digits - those in the right-hand columns - such as 1/3, 1/6, 1/7.
   These are called recurring (or repeating) decimal fractions. These recurring fractions are of two kinds:
   1. Some decimal fractions are solely a collection of digits that repeat from the beginning, such as which is just 3 repeating for everand 1/7 which is endlessly repeated.
      These are called purely repeating decimal fractions
   2. Others start off with a fixed set of digits before they too eventually settle down to an endless repetition of the same digits, such as

      1/6 =

      which begins and then cycles 6 indefinitely

      1/12 =

      which starts before it too starts to repeat 3 for ever.

      These are also called mixed recurring decimal fractions.

At first it is surprising that every fraction fits into one of these two categories:

We could simplify this even further by saying that all terminating decimals end with the infinite cycle of so that every proper fraction is a recurring decimal !

Patterns in recurring decimals

If we take all the fractions with the same denominator, that is, the lower number in a fraction,we can find some amazing patterns too. The first and simplest are the sevenths, the ninths and theelevenths:

Notation for the recurring part of a decimal fraction

Mathematicians use one of two common notations toindicate which of the digits in a decimal fraction are in the repeating part:the period or cycle.

1. a dot is put over the first and last digits in the recurring sequence
   This notation goes back at least to Robertson ().
2. a line is drawn over the repeating part

Both of these are a little awkward on web pages and as the output fromcomputer programs and calculators, so an alternative is also used:

1. enclose the recurring part inside [ and ] brackets

For example:

Things to do

Alternative forms for some recurring fractions

We have just seen that there is no unique way for writing a recurring fractionsince 0.[] = [] = [] =
We could choose one of two conventions: use the longest fixed part or use the shortest fixed part.
For our number example here, it makes sense to write it as 0.[] but in other cases wherethe period begins with 0 we have other choices.
For instance, ¹/₁₁ = .

• With the shortest fixed part (this is the usual form), we have:¹/₁₁ = 0.[09] which makes it a purely recurring decimal fraction.
• Others (such as used in the Mathematica as results from the RealDigits function) prefer to use the longest fixed part and to start a recurring part with a non-zero digit:
  ¹/₁₁ = [90], making this a mixed recurring decimal fraction.

On this page, we use the shortest fixed part for all recurring decimals.

People have suggested that all fractions are recurring onesbecause they all end with or they can end with .And anyway, is the same as = or not?
The answer is "Yes, they are the same!" but here is a longer explanation if you need moreconvincing

Things to do

Fraction to Decimal Calculator

Converting a decimal fraction into a proper fraction

Converting a terminating decimal to a fraction

First let's look at a method of converting a decimal fraction to a proper fraction by hand. This illustrates the maths behind the manicapital.com we can explore with an online calculator to do it for us!
For example,

The process is to write the fraction as a whole number divided by 10^{the number of decimal places} and thensimplify this fraction until it is in its lowest form.

Converting a periodic fraction to a fraction

If the decimal fraction is periodic then it never ends and we need a different approach.

A purely periodic decimal fraction

First, let's take a purely periodic fraction such as 0.[] = 0. Let's call our decimal fraction d.
First we multiply dby 10^{the length of the PERIOD} to make the fractional part of the decimal (to the right of the decimal point)the same as the original number. For this example, we would use 10³ since the period is of length 3: Now, if we subtract the original number form this the part to the right of the decimal point will disappear:The process is to multiply the purely periodic decimal by 10^{length of the period}, subtract the original from this and then divide to form a fraction.

A mixed periodic decimal fraction

If the decimal fraction is mixed, we use a combination of the terminating and purely periodic methods.
For example, [] = Let's call this m.
First, multiply m by 10^{length of the FIXED part} to get a purely periodic fraction. The power to use here is 2:
Then use the purely periodic fraction method above on this value to find m as a fraction: Now find a proper fraction that is the value of m:The calculator in the next section does all the arithmetic using the same methods as above.

Decimal to Fraction Calculator

You can input a decimal fraction and the calculator will convert itto a proper fraction. Give the fixed part (if it has one) followed by the recurring part (if your decimal fraction is not a terminating one).

• If your decimal fraction is a terminating fraction such as , type the digits into the part, leaving the part empty.
• If your decimal fraction is purely recurring as in , type the digits that repeat (in this example it is just ) into the part leaving the part empty. Don't forget to type in all initial and trailing zeroes.
• If your decimal fraction starts with a fixed part and is followed by a recurring part, as in then put into the fixed part and type into the box for the recurring part.

THEN press the button to see the conversionin the RESULTS area.

Things to do

Which decimal fractions terminate and which recur?

How to tell if a fraction is terminating as a decimal fraction

Some experimentation will help answer this question but we will also justify our answers mathematically.
Use the Fraction to Decimal calculator aboveto find all the fractions which have a denominator less than 50 and whose decimal fractions terminate.

Your list should begin with 2, 4, 5, 8, 10, 16, 20 because:

¹/₂ = 0·5, ¹/₄ = 0·25, ¹/₅ = 0·2, ¹/₈ = 0·, ¹/₁₀ = 0·1, ¹/₁₆ = 0·, ¹/₂₀ = 0·05

All these fractions stop after a limited number of digits: they terminate.

Can you spot the pattern that is common to the numbers 2, 4, 5, 8, 10, 16, 20, ?
Try looking at their prime factors.

See if you can find some more numbers, N, whose unit fractions, 1/N terminate and then check your list withA

How to find the length of a terminating decimal fraction

If the proper fraction ⁿ/_d is in its lowest terms thenit terminates if and only if d is divisible by 2or by 5 or by both and is not divisible by any other prime number.
This is because in the process of dividing dinto n by long division, some power of 10 will then be an exact multiple of the denominator and so will leave no remainder andour process of long-division will stop.
We can write this as d = 2^α × 5^βwhere α and β are whole numbers but may be 0.
The decimal fraction for ⁿ/_d has LCM(α, β) decimal placeswhere the LCM(α, β) function means the lowest common multiple of both α and β; the smallest number that isa multiple of both α and β.

The list of denominators with terminating decimals begins:

How to tell if a fraction is purely recurring as a decimal fraction

The rule for a fraction which is purely recurring, that is, its period starts immediately after the decimalpoint, is just the opposite of the rule for detecting a terminating fraction above:The first column is 3, 7, 9, 11, 13, 17, Awhich can be described in several equivalent ways as:

• the numbers coprime to 10
• the numbers with no prime factor in common with 10
• the numbers n for which GCD(10,n) = 1
• the numbers which are not divisible by either 2 or by 5

The third column is the lengths of all the purely periodic decimal fractions:1, 6, 1, 2, 6, 16, 18, 6, AThere is no explicit formula for this series but we can describe it with more detail as we see in the next secion.

Did you notice that all the primes numbers are in this list, except of course 2and 5?
Also, the length of the period of 1/n when n is sucha prime is sometimes n–1 and sometimes not! Can you spot any patterns in theperiod length for a prime n?

The length of the period of a purely periodic decimal fraction

When we divide n into 1 and we eventually reach a remainder of 1 again, then the decimalfraction will repeat. At this point, for example, we have the following if we are using long division tofind the decimal fraction for 1/The division of course is equally correct if we ignored the decimal point so that we would be calculating just with whole manicapital.com that case, we have found that 1 followed by six zeroes, 10⁶, when divided by 21, leaves a remainder of 1. That is:
10⁶ = × 21 + 1 or:

Here we have found that

and that 6 is the smallest power of 10 that leaves a remainder of 1 when divided by

The same is true for all fractions 1/n which have purely periodicdecimal fractions.
Often in textbooks you will see the Greek letter λ (lambda) used for this length:

The list of denominators of purely periodic fractions begins:

Mixed Recurring Fractions

All fractions as a decimal will either terminate or recurr. Some recurr from the first decimal place and are thepurely recurring decimal fractions of the previous section. Others have a few decimal digits at the start and then they recurr. These are the mixed recurring decimal fractions that we look at in this section.
For the fraction n/d, in its lowest terms, to terminate have seen thatthe prime factors of dmust also be prime factors of 10, namely 2 or 5 or both.
Examples are: 2, 8 = 2×2×2, 40 = 2×2×2×5, = 2×2×5×5

For it to be purely recurring, none of the prime factors of dmust be a prime factor of
Examples are: 3, 7, 9 = 3×3, 11 and all the other prime numbers bigger than 5.

So that leaves the mixed recurring decimal fractions which have at least one prime factor incommon with those of 10and at least one not in common withthose of 10
Examples are 6 = 2×3, 15 = 5×3, 30 = 2×5×3
These denominators are of the form

where the p_i are prime numbers other than 2 and 5and α₁ or α₂ can be 0 (which effectively excludes that prime as a factor).

How long are the fixed and recurring parts of a mixed recurring fraction?

Using the above description of the denominator d, we separate its factorization into two parts, so that D = 2^α₁ 5^α₂ and the other primes and their powers are all included in E = d / D

The size of the fixed and recurring parts are determined solely by D.1/Dwill be therefore be a terminating decimal fraction and the number of decimal places it has is the size of the initial fixed part ofreduced fractions with d as the denominator.
The size of the recurring part is determined by the rest of the factorization of the denominator, E.
1/E will be purely recurring and the length of its period is the same as the length of the recurring partof 1/d.

Let's look at some examples:

• 1/12
  12 = 2² × 3 so D = 2², E = 3
  The fixed part is determined by D = 2² = 4 and 1/4 = has 2 decimal digits.
  The recurring part is determined by E=12/4=3 and 1/3 = 0.[3] has a recurring period of length 1. We see that 1/12 = [3] and does indeed have a fixed part of 2 digits and a recurring part of 1 digit.
• 1/38
  38 = 2 × 19 so D = 2, E = 19
  The fixed part is the same length as 1/D = 1/2 = , 1 digit.
  The recurring part is the same length as 1/E = 1/19 = 0.[], 18 digits.
  Check: 1/38 = [] does have a fixed part of 1 digit and a recurring part of 18 digits.
• 1/
  = 2 × 5³ × 7 × 11 so D = 2 × 5³, E = 7 × 11
  The fixed part is given by D = 2 × 5³ = and 1/ = has 3 decimal digits.
  The recurring part is determined by the rest of the factorisation: E = 7 × 11 = 77 and 1/77 = 0.[] has a recurring period of length 6.
  We find that 1/ = [] and does indeed have a fixed part of 3 digits and a recurring part of 6 digits.

Below there is a calculator which will find denominators or reduced fractions for given lengths of fixed and/or recurring parts.

The denominators of fractions with purely recurring decimals are:

The denominators that have decimals that eventually recur, either as purely recurring decimals or else as mixed recurring decimals arethose which do not have a terminating decimal expansion:

Fractions with the same denominator

We have mainly looked at ¹/_n up to nowand how many decimal places it has and if it is periodic or manicapital.com can we say about ^k/_n?
Some examples are:

n n/5 Decimal 1 1/5 2 2/5 3 3/5 4 4/5

n n/6 Decimal 1 1/6 [6] 2 1/3 0.[3] 3 1/2 4 2/3 0.[6] 5 5/6 [3]

n n/8 Decimal 1 1/8 2 1/4 3 3/8 4 1/2 5 5/8 6 3/4 7 7/8

n n/12 Decimal 1 1/12 [3] 2 1/6 [6] 3 1/4 4 1/3 0.[3] 5 5/12 [6] 6 1/2 n n/12 Decimal 7 7/12 [3] 8 2/3 0.[6] 9 3/4 10 5/6 [3] 11 11/12 [6]

It looks as if the numerator does not matter but that it is only the denominator of the fraction in its lowest termsthat matters - and indeed this proves correct!
The character of the decimal fraction -whether it terminates or not and the lengths of the fixed and recurring parts - is solely determined by thedenominator provided the fraction is in its lowest terms.

The rules for decimal (base 10) fraction for ¹/_d in its lowest form

Here are the mathematical details (optional) to show why the numerator does not matter:

For example:
¹³/ has an initial fixed part of 2 decimal places and a periodic repeating part of 3 dps and so will all decimal fractions for ⁿ/provided that n has no factor in common with :

n/ lowest terms Decimal 1/ 1/ [] 2/ 1/54 [] 3/ 1/36 [7] 4/ 1/27 0.[] 5/ 5/ [] 6/ 1/18 [5] 7/ 7/ [] 8/ 2/27 0.[] 9/ 1/12 [3] 10/ 5/54 []

n/ lowest terms Decimal 11/ 11/ [] 12/ 1/9 0.[1] 13/ 13/ [] 14/ 7/54 [] 15/ 5/36 [8] 16/ 4/27 0.[] 17/ 17/ [] 18/ 1/6 [6] 19/ 19/ [] 20/ 5/27 0.[]

n/ lowest terms Decimal 21/ 7/36 [4] 22/ 11/54 [] 23/ 23/ [] 24/ 2/9 0.[2] 25/ 25/ [] 26/ 13/54 [] 27/ 1/4 28/ 7/27 0.[] 29/ 29/ [] 30/ 5/18 [7]

Same denominator Decimal Fraction Calculator

This calculator finds all those fractions with the same denominator and counts them, displays all the numeratorsor shows the all the decimal places or the period. The denominator can be a number up to 16 digits long and theCount is displayed but you will need to confirm you want to see all the numerators or decimals if there are more than 50numerators.

The number of fractions with the same denominator and Euler's Totient function φ(n)

The Calculator above computes the number of fractions with the same denominator by using a relatively simple formulawhich is why it is so fast. The number of fractions with the same denominator dis an important function in mathematics andin Number Theory in particular. Eulerused it in his research and called it φ(d), the (Euler) Totient function. It has nothing to do with the goldenmean number φ as that is a completely different use for the Greek letter phi! Here is a table of the number of proper (reduced) fractions with a given denominator dor φ(d):

d	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20
φ(d)	1	2	2	4	2	6	4	6	4	10	4	12	6	8	8	16	6	18	8

After φ(2)=1, all the values are even.
This is the series A, the totient numbers and it plays an important part in many results in Number Theory.

Nontotient Numbers

All values of φ(n) are even for n>2. But not alleven numbers are the result of φ(n) for some n.
In other words when we list the counts of proper fractions with a given denominator, there are some counts thatnever appear in the list and these are the nontotient numbers.For instance,
we can find n for φ(n) = 12, sinceφ(13) = 12 and there are 12 proper fractions less than 1 with a denominator of
Similarly, φ(n) = 16 sinceφ(17) = 16 and there are 16 proper fractions less than 1 with a denominator of
But there is no solution to φ(n) = there is no denominator that has exactly14 reduced fractions!
14 is the smallest nontotient number.
The list begins:

Fractions with the same lengths of period, fixed part or terminating part

We have seen how to compute the length of the terminating part of of the fixed and periodic parts of a given fraction. Now let's ask the question in reverse:

Terminating Decimals of a given length

A terminating decimal fraction ⁿ/_d must have d = 2^α × 5^β ( α and β may be 0) and the length of its (terminating) decimal is LCM(α, β) which we saw above.
So, given a length t of a terminating decimal fraction we need to find α and β with LCM(α, β) = t. This means that both α and β are among the factors of t or else one of them is 0 and the other is t.
For example, if t = 2, we have As a check, there are 99 decimals with just 2 decimal places, from to , so the number of fractions in their lowest terms with the above denominators must sum to 99, you might think but this is wrong! Why? Because is not a terminating fraction of 2 digits but just 1.
We must exclude those 2-digit numbers that end in 0, leaving just 90 terminating fractions of 2 digits (decimal places).

Purely periodic decimals of a given length

A common notation used for the length of the period of the decimal for 1/dis the Greek lowercase letter for &#; for length, which is lambdaλ.
λ will be the smallest power of 10 that has a remainder of 1 when divided by manicapital.com we are given a value for λ we want to find the d for which10^λ = 1 (mod d) which means that d will be a factor of 10^λ - 1.
Here are the factors of 10^λ - 1 for various values of λ = length of period:Some interesting facts arise from this table:

• If we find one of the factors d in row λ but not in any earlier row, then 1/d is purely periodic with a period of length λ.
• Any other fraction n/d with GCD(n,d)=1 will also have a period of the same length.
• The number of divisors of 10ⁿ–1 varies considerably (A and this table includes the factor 1):

  n	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20
  #divisors	3	6	8	12	12	64	12	48	20	48	12		24	48			12		6

• For instance,
  10²³–1 has only 6 factors but
  10²⁴–1 has
  10³⁰–1 has
  Here are three consecutive values of the power with wildly differing numbers of factors:
  10⁵⁹–1 has 12 factors but
  10⁶⁰–1 has and
  10⁶¹–1 has
• This is because if m is a factor of n then 10^m–1 is a factor of 10ⁿ–1.
  So numbers n that have many factors will give rise to many factors for 10ⁿ–1
• All numbers 10ⁿ–1 for n>1 have at least these 6 factors: 1, 3, 9, (10ⁿ–1)/9 = , (10ⁿ–1)/3 = , (10ⁿ–1)/9 =
• These 6 factors are the only divisors for 10²–1, 10¹⁹–1 and 10²³–1.
• So, for n = 2, 19 and 23 the larger 3 factors will be their only denominators for purely recurring fractions with those period lengths.

Decimals of given fixed part length and period length

Let's call the fixed part length t and the period length &#and either may be 0.
Combining the results of the above two sections, we find all numbers Twith terminating decimal of length t and all numbers Rwhich are purely recurring with a period of length λ, as described above.
The mixed fraction denominators which have both fixed part length tand a period of length λ are those numbers made from a product of one of the T with one of the R.
From the sections above we see that

• the denominators of fractions whose decimal is only a fixed-part of F decimals are factors of 10^F
• the denominators of fractions whose decimal is only a period of P decimals are factors of 10^P−1
• the denominators of fractions whose decimal has both a fixed part and periodic part are a product of the denominators with just the fixed part and just the periodic part

Similar sized Decimals Calculator

The fixed length and period length cannot be larger than 15 with this calculator. Leave the 'up to'value empty to include all valid integers with up to 16 digits.
This Calculator finds the number of distinct denominators that can occur for decimal numbers with a given number of fixed-part decimal places and a given numberof periodic-part decimal places.
For example, there are 9 distinct decimals of the form 0.d with 1 fixed decimal digit d>0 and no periodic decimals as follows:but there are only 3 possible denominators for these fractions: 2, 5 and

Plots of Period lengths

Here are plots of the lengths of the periods of the purely periodic fractions 1/N for different ranges of N:

Decimal fractions that look like special sequences

Now we look at some special kinds of decimal fractions, ones that contain a whole special series in their decimals!It is amazing to find that some fractions have a decimal expansion with a special patternsuch as the powers of 2 as in which is the first few decimal places of 1/98!Can you guess which simple fraction begins with the powers of 3: 03 09 27 ?
How about ?

However after a few more decimal places the series often get "muddled" and the pattern seems to disappear. This section unravels some of the simple maths behind this mystery.

Powers of a number

If we look more closely at some simple fractions near 1/we see their decimal fractions are special:

1/99	= 0. 01 01 01 01 01 01 01 01 01 1
1/98	= 0. 01 02 04 08 16 32 65 30 61 2
1/97	= 0. 01 03 09 27 83 50 51 54 63 9

But 1/98 looks like 01 02 04 08 16 32 the powers of two and 98 = - 2.
and 1/97 looks like 01 03 09 27 8(1) the powers of three and 97 = - 3.
In fact, even 1/99 fits this pattern because 99 = - 1 and it would then be the powers of 1, that is 01 01 01 01 01 which it is!
But soon the pattern of powers disappears. Why?
Because it is really there, just masked as the powers get larger

Shift-and-add applied to the a series makes the fraction!

We take the powers of 2 but moving each power exactly 2 places to the right each time and then we add them:and we see this is indeed the start of 1/98 as a decimal.

Did you notice that

• the powers of 2 taken two digits at a time are in the fraction 1/(10² - 2) = 1/98
• the powers of 3 taken two digits at a time are in the fraction 1/(10² - 3) = 1/97

Any guesses as to a fraction whose decimals are the powers of 2 using three digits for each?

Things to do

So we observe (or guess!) that

Editors

The OAI Executive:
Carl Lagoze <lagoze@manicapital.com > -- Cornell University - Computer Science
Herbert Van de Sompel <herbertv@manicapital.com > -- Los Alamos National Laboratory - Research Library

From the OAI Technical Committee:
Michael Nelson <manicapital.com@manicapital.com > -- NASA - Langley Research Center
Simeon Warner <simeon@manicapital.com > -- Cornell University - Computer Science

Table of Contents

1. Introduction
2. Definitions and Concepts
    Harvester
    Repository
    Item
    Unique Identifier
    Record
       Deleted records
    Set
    Selective Harvesting
       Selective Harvesting and Datestamps
       Selective Harvesting and Sets
3. Protocol Features
    HTTP Embedding of OAI-PMH requests
       HTTP Request Format
       HTTP Response Format
       Response Compression
    XML Response Format
       XML Schema for Validating Responses to OAI-PMH Requests
    UTCdatetime
       UTCdatetime in Protocol Requests
       UTCdatetime in Protocol Responses
    metadataPrefix and Metadata Schema
    Flow Control
       Idempotency of resumptionTokens
    Error and Exception Conditions
4. Protocol Requests and Responses
    GetRecord
    Identify
    ListIdentifiers
    ListMetadataFormats
    ListRecords
    ListSets
5. Dublin Core
6. Implementation Guidelines
Acknowledgements
Document History

The Open Archives Initiative Protocol for Metadata Harvesting (referred to as the OAI-PMH in the remainder of this document) provides an application-independent interoperability framework based on metadata harvesting. There are two classes of participants in the OAI-PMH framework:

• Data Providers administer systems that support the OAI-PMH as a means of exposing metadata; and
• Service Providers use metadata harvested via the OAI-PMH as a basis for building value-added services.

In this document the key words "must", "must not", " required", "shall", "shall not", "should", " should not", "recommended", "may", and "optional " in bold face are to be interpreted as described in RFC . An implementation is not conformant if it fails to satisfy one or more of the "must" or "required" level requirements for the protocols it implements.

This document refers in several places to "community-specific" practices to which individual protocol implementations may conform. These practices are described in an accompanying Implementation Guidelines document.

Harvester

A harvester is a client application that issues OAI-PMH requests. A harvester is operated by a service provider as a means of collecting metadata from repositories.

Repository

A repository is a network accessible server that can process the 6 OAI-PMH requests in the manner described in this document. A repository is managed by a data provider to expose metadata to harvesters. To allow various repository configurations, the OAI-PMH distinguishes between three distinct entities related to the metadata made accessible by the OAI-PMH.

• resource - A resource is the object or "stuff" that metadata is "about". The nature of a resource, whether it is physical or digital, or whether it is stored in the repository or is a constituent of another database, is outside the scope of the OAI-PMH.
• item - An item is a constituent of a repository from which metadata about a resource can be disseminated. That metadata may be disseminated on-the-fly from the associated resource, cross-walked from some canonical form, actually stored in the repository, etc.
• record - A record is metadata in a specific metadata format. A record is returned as an XML-encoded byte stream in response to a protocol request to disseminate a specific metadata format from a constituent item.

Item

An item is a constituent of a repository from which metadata about a resource can be disseminated. An item is conceptually a container that stores or dynamically generates metadata about a single resource in multiple formats, each of which can be harvested as records via the OAI-PMH. Each item has an identifier that is unique within the scope of the repository of which it is a constituent.

Unique Identifier

A unique identifier unambiguously identifies an item within a repository; the unique identifier is used in OAI-PMH requests for extracting metadata from the item. Items may contain metadata in multiple formats. The unique identifier maps to the item, and all possible records available from a single item share the same unique identifier.

The format of the unique identifier must correspond to that of the URI (Uniform Resource Identifier) syntax. Individual communities may develop community-specific URI schemes for coordinated use across repositories. The scheme component of the unique identifiers must not correspond to that of a recognized URI scheme unless the identifiers conform to that scheme. Repositories may implement the syntax described in the accompanying Implementation Guidelines document.

Unique identifiers play two roles in the protocol:

1. Response: Identifiers are returned by both the and requests.
2. Request: An identifier, in combination with a , is used in the request as a means of requesting a record in a specific metadata format from an item.

Note that the identifier described here is not that of a resource. The nature of a resource identifier is outside the scope of the OAI-PMH. To facilitate access to the resource associated with harvested metadata, repositories should use an element in metadata records to establish a linkage between the record (and the identifier of its item) and the identifier (URL, URN, DOI, etc.) of the associated resource. The mandatory Dublin Core format provides the element that should be used for this purpose.

Record

A record is metadata expressed in a single format. A record is returned in an XML-encoded byte stream in response to an OAI-PMH request for metadata from an item. A record is identified unambiguously by the combination of the unique identifier of the item from which the record is available, the identifying the metadata format of the record, and the datestamp of the record. The XML-encoding of records is organized into the following parts:

• header -- contains the unique identifier of the item and properties necessary for selective harvesting. The header consists of the following parts:
  • the unique identifier -- the unique identifier of an item in a repository;
  • the datestamp -- the date of creation, modification or deletion of the record for the purpose of selective harvesting.
  • zero or more setSpec elements -- the set membership of the item for the purpose of selective harvesting.
  • an optional attribute with a value of indicates the withdrawal of availability of the specified metadata format for the item, dependent on the repository support for deletions.
• metadata -- a single manifestation of the metadata from an item. The OAI-PMH supports items with multiple manifestations (formats) of metadata. At a minimum, repositories must be able to return records with metadata expressed in the Dublin Core format, without any qualification. Optionally, a repository may also disseminate other formats of metadata. The specific metadata format of the record to be disseminated is specified by means of an argument -- the -- in the or request that produces the record. The request returns the list of all metadata formats available from a repository, or for a specific item (which can be specified as an argument to the request).
• about -- an optional and repeatable container to hold data about the metadata part of the record. The contents of an about container must conform to an XML Schema. Individual implementation communities may create XML Schema that define specific uses for the contents of about containers. Two common uses of about containers are:
  • rights statements: some repositories may find it desirable to attach terms of use to the metadata they make available through the OAI-PMH. No specific set of XML tags for rights expression is defined by OAI-PMH, but the about container is provided to allow for encapsulating community-defined rights tags.
  • provenance statements: One suggested use of the about container is to indicate the provenance of a metadata record, e.g. whether it has been harvested itself and if so from which repository, and when. An XML Schema for such a provenance container, as well as some supporting information is available from the accompanying Implementation Guidelines document.

The following example shows an XML-encoding of a record and its components:

• the header part with:
  • a unique identifier of the item from which the record was disseminated, equal to oai:manicapital.com:cs/;
  • the datestamp of the record equal to ;
  • two setSpecs, respectively and , indicating that the item from which the record was disseminated belongs to two sets of the repository;
• the metadata part. This consists of a single root tag - in the example the tag - with the nested tags belonging to the corresponding metadata format - in the example, Dublin Core elements such as . Note that the root tag within the metadata part includes a number of attributes that are common to all XML documents that use namespaces and schema validity:
  • namespace declarations -- the declarations of the namespaces used within the metadata part, each of which is prefixed with . Namespace declarations within the metadata part fall into two categories:
    • metadata format specific namespace(s) - every metadata part must include one or more prefixed attributes that define the correspondence between a metadata format prefix -- e.g. -- and the namespace URI (as defined by the XML namespace specification ) of the respective metadata format. Some metadata formats employ tags from multiple namespaces, requiring multiple prefixed attributes -- in the example, there are declarations for both and .
    • xml schema namespace - every metadata part must include the attribute , the value of which must always be the URI shown in the example, which is the namespace URI for XML schema.
  • -- the value of which is a URI, URL pair; the first is the namespace URI (as defined by the XML namespace specification) of the metadata that follows in this part, and the second is the URL of the XML schema for validation of the metadata that follows.
• one about part of the record which uses the _ schema, described in the accompanying Implementation Guidelines document, as a means to provide information regarding the origins of the metadata part of the record. Note that the root element within each about part has the same structure as the root element in the metadata part.

<header> <identifier>oai:arXiv:cs/</identifier> <datestamp></datestamp> <setSpec>cs</setSpec> <setSpec>math</setSpec> </header> <metadata> <oai_dc:dc xmlns:oai_dc="manicapital.com" xmlns:dc="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <dc:title>Using Structural Metadata to Localize Experience of Digital Content</dc:title> <dc:creator>Dushay, Naomi</dc:creator> <dc:subject>Digital Libraries</dc:subject> <dc:description>With the increasing technical sophistication of both information consumers and providers, there is increasing demand for more meaningful experiences of digital information. We present a framework that separates digital object experience, or rendering, from digital object storage and manipulation, so the rendering can be tailored to particular communities of users. </dc:description> <dc:description>Comment: 23 pages including 2 appendices, 8 figures</dc:description> <dc:date></dc:date> <dc:type>e-print</dc:type> <dc:identifier>manicapital.com</dc:identifier> </oai_dc:dc> </metadata> <about> <provenance xmlns="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <originDescription harvestDate="TZ" altered="true"> <baseURL>manicapital.com</baseURL> <identifier>oai:r2:klik</identifier> <datestamp></datestamp> <metadataNamespace>manicapital.com</metadataNamespace> </originDescription> </provenance> </about>

Deleted records

If a record is no longer available then it is said to be deleted. Repositories must declare one of three levels of support for deleted records in the element of the Identify response:

• - the repository does not maintain information about deletions. A repository that indicates this level of support must not reveal a deleted status in any response.
• - the repository maintains information about deletions with no time limit. A repository that indicates this level of support must persistently keep track of the full history of deletions and consistently reveal the status of a deleted record over time.
• - the repository does not guarantee that a list of deletions is maintained persistently or consistently. A repository that indicates this level of support may reveal a deleted status for records.

If a repository does not keep track of deletions then such records will simply vanish from responses and there will be no way for a harvester to discover deletions through continued incremental harvesting. If a repository does keep track of deletions then the datestamp of the deleted record must be the date and time that it was deleted. Responses to request for a deleted record must then include a with the attribute , and must not include or parts. Similarly, responses to selective harvesting requests with set membership and date range criteria that include deleted records must include the headers of these records. Incremental harvesting will thus discover deletions from repositories that keep track of them.

Deleted status is a property of individual records. Like a normal record, a deleted record is identified by a unique identifier, a metadataPrefix and a datestamp. Other records, with different metadataPrefix but the same unique identifier, may remain available for the item.

Set

A set is an optional construct for grouping items for the purpose of selective harvesting. Repositories may organize items into sets. Set organization may be flat, i.e. a simple list, or hierarchical. Multiple hierarchies with distinct, independent top-level nodes are allowed. Hierarchical organization of sets is expressed in the syntax of the parameter as described below. When a repository defines a set organization it must include set membership information in the headers of items returned in response to the , and requests.

Each node in a set organization of a repository has:

• a -- a colon [] separated list indicating the path from the root of the set hierarchy to the respective node. Each element in the list is a string consisting of any valid URI unreserved characters, which must not contain any colons []. Since a forms a unique identifier for the set within the repository, it must be unique for each set. Flat set organizations have only sets with that do not contain any colons [].
• a -- a short human-readable string naming the set.
• a -- an optional and repeatable container that may hold community-specific XML-encoded data about the set; the accompanying Implementation Guidelines document provides suggestions regarding the usage of this container.

The following is an example of a possible set hierarchy in a repository:

• Institutions
  • Oceanside University of Nebraska
  • Valley View University of Florida
• Subjects
  • Existential Kenesiology
  • Quantum Psychology

The following table shows a possible representation of the above set hierarchy by means of and respective values.

Institutions

Oceanside University of Nebraska

Valley View University of Florida

Subjects

Existential Kenesiology

Quantum Psychology

An item may be organized in one set, several sets, or no sets at all. In the example above, it is conceivable that an individual item is organized in both and . A harvester should not assume that harvesting every set in a repository will retrieve metadata from all items in the repository. Items may also be assigned to interior nodes in the set hierarchy.

The actual meaning of a set or of the arrangement of sets in a repository is not defined in the protocol. It is expected that individual communities may formulate well-defined set configurations with perhaps a controlled vocabulary for and , and may even develop mechanisms for exposing these to harvesters. For example, a group of cooperating e-print archives in a specific discipline may agree on sets that arrange metadata in their repositories based on a controlled subject classification.

A repository's set hierarchy is represented in the protocol via . returns a list indicating the configuration of sets in a repository. Each member of this list must include a and a and may include a . and requests may include an optional argument, the value of which is a , to specify the target set for selective harvesting. In the previous example of a set hierarchy, the could be used in a request to return only those records that are disseminated from items organized in the set represented by this . Five issues should be noted here:

• If a repository supports sets then it must include set membership information in response to , and requests. The list of elements should include only the minimum number of elements required to specify the set membership. Using the previous example of a set hierarchy, the header for an item organized in set should not include since that is implied by the .
• An item may be organized in more than one set; meaning that different arguments may return the same record(s).
• An item need not be organized in any set; meaning that an exhaustive repetition of requests with all possible is not guaranteed to return all records in the repository. The only guaranteed methods of harvesting all records or headers are or requests with no argument.
• When a is used as an argument, the response must include records or headers from all items in the set specified by the , and all records or headers from items in sets that are descendant from the specified set. Using the previous example of a set hierarchy, a of to the request will return all records from metadata organized within the set with a value equal to and within the descendent sets with setSpec values equal to and .
• The set hierarchy of a repository may include sets that are empty.

Selective Harvesting

Selective harvesting allows harvesters to limit harvest requests to portions of the metadata available from a repository. The OAI-PMH supports selective harvesting with two types of harvesting criteria that may be combined in an OAI-PMH request: datestamps and set membership.

Selective Harvesting and Datestamps

Harvesters may use datestamps to harvest only those records that were created, deleted, or modified within a specified date range. To specify datestamp-based selective harvesting, datestamps are included as values of the optional arguments, and , in the and requests. Harvesting is restricted to the range specified by the and arguments, extending back to the earliest datestamp if is omitted, and forward to the most recent datestamp if is omitted. Range limits are inclusive: specifies a bound that must be interpreted as "greater than or equal to", specifies a bound that must be interpreted as "less than or equal to". Therefore, the argument must be less than or equal to the argument. Otherwise, a repository must issue a error.

Repositories must support selective harvesting with the and arguments expressed at day granularity. Optional support for seconds granularity is indicated in the response to the request. The value of datestamps in both requests and responses must comply to the specifications for UTCdatetime in this document. A repository must update the datestamp of a record if a change occurs, the result of which would be a change to the metadata part of the XML-encoding of the record. Such changes include, but are not limited to, changes to the metadata of the record, changes to the metadata format of the record, introduction of a new metadata format, termination of support for a metadata format, etc.

Datestamp ranges for selective harvesting are expressed in the and arguments that may be submitted in the and requests. Repositories must use the following rules to create a response matching the specified datestamp range according to the type of change that occurred within the repository. The response to a request follows the same rules but is abbreviated to include only headers rather than records.

• modification - the response must include records, corresponding to the argument, which have changed within the bounds of the and arguments.
• creation - the response must include records, corresponding to the argument, that have become available from the repository within the bounds of the and arguments.
• deletion - depending on the level at which a repository keeps track of deleted records, the response may include headers of records, corresponding to the argument, which have been withdrawn from the repository within the bounds of the and arguments. Deleted status is indicated via the status attribute of the header element and no metadata is included.

Every header returned by the , or requests contains a datestamp, which reflects the most recent date and time of the creation, modification, or deletion according to the rules defined above.

Selective Harvesting and Sets

Harvesters may specify set membership as a criteria for selective harvesting. To specify set-based selective harvesting, a is included as the value of the optional argument to the and requests, thereby specifying selective harvesting of records from items within the respective set.

When a is used as an argument, the response must include:

• the records corresponding to the argument, or headers thereof in the case of deleted records, available from those items in the set specified by the
• the records corresponding to the argument, or headers thereof in the case of deleted records, available from those items in sets that are descendant from the specified set.

HTTP Embedding of OAI-PMH requests

OAI-PMH requests are expressed as HTTP requests. A typical implementation uses a standard Web server that is configured to dispatch OAI-PMH requests to the software handling these requests. The remainder of this section describes the aspects of the protocol that are specific to the HTTP embedding.

HTTP Request Format

OAI-PMH requests must be submitted using either the HTTP or methods. has the advantage of imposing no limitations on the length of arguments. Repositories must support both the and methods. There is a single base URL for all requests. The base URL specifies the Internet host and port, and optionally a path, of an HTTP server acting as a repository. Repositories expose their base URL as the value of the element in the response. Note that the composition of any path is determined by the configuration of the repository's HTTP server.

In addition to the base URL, all requests consist of a list of keyword arguments, which take the form of pairs. Arguments may appear in any order and multiple arguments must be separated by ampersands []. Each OAI-PMH request must have at least one pair that specifies the OAI-PMH request issued by the harvester:

The number and nature of additional pairs depends on the arguments for the individual request.

Encoding an OAI-PMH request in a URL for an HTTP GET

URLs for requests have keyword arguments appended to the base URL, separated from it by a question mark []. For example, the URL of a request to a repository with base URL that is might be:

However, since special characters in URIs must be encoded, the correct form of the above request URL is:

Encoding an OAI-PMH request in an HTTP POST

Keyword arguments are carried in the message body of the HTTP . The of the request must be . For example, submitting the same request as above using the method would use just the base URL as the URL, with the format of the being:

Encoding of special characters in keyword arguments of OAI-PMH requests

The syntax rules for URIs restrict a few characters to special roles in certain contexts, and require that if these characters are used in any other way that they must be written as an escape sequence, i.e. a percent sign followed by the character code in hexadecimal. The reserved characters include:

Character	URI Role	Escape Sequence
Path Component Separator
Query Component Separator
Fragment Identifier
Name/Value Separator
Argument Separator in Query Component
Host Port Separator
Authority Namespace Separator
	Space Character
%	Escape Indicator
+	Escaped Space

As a result, these characters must be represented by their respective escape sequence if their use does not correspond to their established URI role. In case of the OAI-PMH, this means that the reserved characters must be encoded when they appear in the part of the pairs of the request. This applies for both the and encoding of the OAI-PMH requests.

HTTP Response Format

Responses to requests are formatted as HTTP responses, with appropriate HTTP header fields.

Content-Type

The returned for all OAI-PMH requests must be .

Status-Code

OAI-PMH errors are distinguished from HTTP . Since OAI-PMH uses HTTP as a transport layer, servers implementing OAI-PMH must conform to HTTP status code definitions and report relevant HTTP transport layer status via those . OAI-PMH repositories may employ HTTP in addition to "". For instance, the following may be useful for load balancing in OAI repositories:

• - Allows the repository to temporarily redirect an OAI-PMH request to another repository. The URI of the temporary repository should be given by the field in the HTTP response.
• - Service unavailable, a period is specified. Harvesters should wait this period before attempting another OAI-PMH request.

Response Compression

Response compression is optional in OAI-PMH. Compression of responses to OAI-PMH requests is handled at the level of HTTP, with the following restrictions:

• Harvesters may include an header in their OAI-PMH requests to specify response compression preferences.
• Harvesters that do not include an header in their requests will always receive uncompressed responses.
• When a request includes an header the list of encodings must include the (no compression) encoding (with a non-zero ).
• Repositories must support the HTTP encoding.
• Repositories should express the encodings they support in addition to by including elements in the response.

XML Response Format

All responses to OAI-PMH requests must be well-formed XML instance documents. Encoding of the XML must use the UTF-8 representation of Unicode. Character references, rather than entity references, must be used. Character references allow XML responses to be treated as stand-alone documents that can be manipulated without dependency on entity declarations external to the document.

The XML data for all responses to OAI-PMH requests must validate against the XML Schema shown at the end of this section . As can be seen from that schema, responses to OAI-PMH requests have the following common markup:

1. The first tag output is an XML declaration where the is always and the is always , eg:
2. The remaining content is enclosed in a root element with the name OAI-PMH. This element must have three attributes that define the XML namespaces used in the remainder of the response and the location of the validating schema:
   • -- the value of which must be the namespace URI of the OAI-PMH ().
   • -- the value of which must be the namespace URI for XML schema ().
   • -- is a pair, the first part of which is the namespace URI (as defined by the XML namespace specification ) of the OAI-PMH (), and the second part is the URL of the XML schema for validation of the response ().
3. For all responses, the first two children of the root element are:
   • -- a UTCdatetime indicating the time and date that the response was sent. This must be expressed in UTC.
   • -- indicating the protocol request that generated this response. The rules for generating the element are as follows:
     • The content of the element must always be the base URL of the protocol request;
     • The only valid attributes for the element are the of the pairs of protocol request. The attribute values must be the corresponding of those pairs;
     • In cases where the request that generated this response did not result in an error or exception condition, the attributes and attribute values of the element must match the pairs of the protocol request;
     • In cases where the request that generated this response resulted in or error condition, the repository must return the base URL of the protocol request only. Attributes must not be provided in these cases.
4. The third child of the root element is either:
   • an element that must be used in case of an error or exception condition;
   • an element with the same name as the verb of the respective OAI-PMH request.

An example of a successful reply to the request shown above is of the form:

<?xml version="" encoding="UTF-8" ?> <OAI-PMH xmlns="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <responseDate>TZ</responseDate> <request verb="GetRecord" identifier="oai:manicapital.com:hep-th/" metadataPrefix="oai_dc">manicapital.com</request> <GetRecord> <record> </record> </GetRecord> </OAI-PMH>

XML Schema for Validating Responses to OAI-PMH Requests

<schema targetNamespace="manicapital.com" xmlns="manicapital.com" xmlns:oai="manicapital.com" elementFormDefault="qualified" attributeFormDefault="unqualified"> <annotation> <documentation> XML Schema which can be used to validate replies to all OAI-PMH v requests. Herbert Van de Sompel, Validated with XML Spy v on Validated with XSV / on Added definition of protocolVersionType instead of using anonymous type. No change of function. Simeon Warner, Tightened definition of UTCdatetimeType to enforce the restriction to UTC Z notation. Simeon Warner, Corrected pattern matches for setSpecType and metadataPrefixType to agree with protocol specification. Simeon Warner, Spelling correction. Simeon Warner, -- define OAI-PMH error conditions --> <!-- =============================== --> <complexType name="OAI-PMHerrorType"> <simpleContent> <extension base="string"> <attribute name="code" type="oai:OAI-PMHerrorcodeType" use="required"/> </extension> </simpleContent> </complexType> <simpleType name="OAI-PMHerrorcodeType"> <restriction base="string"> <enumeration value="cannotDisseminateFormat"/> <enumeration value="idDoesNotExist"/> <enumeration value="badArgument"/> <enumeration value="badVerb"/> <enumeration value="noMetadataFormats"/> <enumeration value="noRecordsMatch"/> <enumeration value="badResumptionToken"/> <enumeration value="noSetHierarchy"/> </restriction> </simpleType> <!-- define OAI-PMH verb containers --> <!-- define basic types used in replies to GetRecord, ListRecords, ListIdentifiers --> <!-- ======================================= --> <complexType name="recordType"> <annotation> <documentation>A record has a header, a metadata part, and an optional about container</documentation> </annotation> <sequence> <element name="header" type="oai:headerType"/> <element name="metadata" type="oai:metadataType" minOccurs="0"/> <element name="about" type="oai:aboutType" minOccurs="0" maxOccurs="unbounded"/> </sequence> </complexType> <complexType name="headerType"> <annotation> <documentation>A header has a unique identifier, a datestamp, and setSpec(s) in case the item from which the record is disseminated belongs to set(s). the header can carry a deleted status indicating that the record is deleted.</documentation> </annotation> <sequence> <element name="identifier" type="oai:identifierType"/> <element name="datestamp" type="oai:UTCdatetimeType"/> <element name="setSpec" type="oai:setSpecType" minOccurs="0" maxOccurs="unbounded"/> </sequence> <attribute name="status" type="oai:statusType" use="optional"/> </complexType> <simpleType name="identifierType"> <restriction base="anyURI"/> </simpleType> <simpleType name="statusType"> <restriction base="string"> <enumeration value="deleted"/> </restriction> </simpleType> <complexType name="metadataType"> <annotation> <documentation>Metadata must be expressed in XML that complies with another XML Schema (namespace=#other). Metadata must be explicitly qualified in the response.</documentation> </annotation> <sequence> <any namespace="##other" processContents="strict"/> </sequence> </complexType> <complexType name="aboutType"> <annotation> <documentation>Data "about" the record must be expressed in XML that is compliant with an XML Schema defined by a community.</documentation> </annotation> <sequence> <any namespace="##other" processContents="strict"/> </sequence> </complexType> <complexType name="resumptionTokenType"> <annotation> <documentation>A resumptionToken may have 3 optional attributes and can be used in ListSets, ListIdentifiers, ListRecords responses.</documentation> </annotation> <simpleContent> <extension base="string"> <attribute name="expirationDate" type="dateTime" use="optional"/> <attribute name="completeListSize" type="positiveInteger" use="optional"/> <attribute name="cursor" type="nonNegativeInteger" use="optional"/> </extension> </simpleContent> </complexType> <complexType name="descriptionType"> <annotation> <documentation>The descriptionType is used for the description element in Identify and for setDescription element in ListSets. Content must be compliant with an XML Schema defined by a community.</documentation> </annotation> <sequence> <any namespace="##other" processContents="strict"/> </sequence> </complexType> <simpleType name="UTCdatetimeType"> <annotation> <documentation>Datestamps are to either day (type date) or to seconds granularity (type oai:UTCdateTimeZType)</documentation> </annotation> <union memberTypes="date oai:UTCdateTimeZType"/> </simpleType> <simpleType name="UTCdateTimeZType"> <restriction base="dateTime"> <pattern value=".*Z"/> </restriction> </simpleType> <!-- define types used for Identify verb only --> <!-- ======================================== --> <simpleType name="protocolVersionType"> <restriction base="string"> <enumeration value=""/> </restriction> </simpleType> <simpleType name="emailType"> <restriction base="string"> <pattern value="\S+@(\S+\.)+\S+"/> </restriction> </simpleType> <simpleType name="deletedRecordType"> <restriction base="string"> <enumeration value="no"/> <enumeration value="persistent"/> <enumeration value="transient"/> </restriction> </simpleType> <simpleType name="granularityType"> <restriction base="string"> <enumeration value="YYYY-MM-DD"/> <enumeration value="YYYY-MM-DDThh:mm:ssZ"/> </restriction> </simpleType> <!-- define types used for ListMetadataFormats verb only --> <!-- =================================================== --> <complexType name="metadataFormatType"> <sequence> <element name="metadataPrefix" type="oai:metadataPrefixType"/> <element name="schema" type="anyURI"/> <element name="metadataNamespace" type="anyURI"/> </sequence> </complexType> <simpleType name="metadataPrefixType"> <restriction base="string"> <pattern value="[A-Za-z\-_\.!~\*'\(\)]+"/> </restriction> </simpleType> <!-- define types used for ListSets verb --> <!-- =================================== --> <complexType name="setType"> <sequence> <element name="setSpec" type="oai:setSpecType"/> <element name="setName" type="string"/> <element name="setDescription" type="oai:descriptionType" minOccurs="0" maxOccurs="unbounded"/> </sequence> </complexType> <simpleType name="setSpecType"> <restriction base="string"> <pattern value="([A-Za-z\-_\.!~\*'\(\)])+(:[A-Za-z\-_\.!~\*'\(\)]+)*"/> </restriction> </simpleType> </schema>

This Schema is available at manicapital.com

UTCdatetime

Dates and times are uniformly encoded using ISO and are expressed in UTC throughout the protocol. When time is included, the special UTC designator ("") must be used. UTC is implied for dates although no timezone designator is specified. For example, is UTC PM on March 20th UTCdatetime is used in both protocol requests and protocol replies, in the way described in the following sections.

UTCdatetime in Protocol Requests

Datestamps used as values of the optional arguments and in the and requests are encoded using ISO and are expressed in UTC. These arguments are used to specify datestamp-based selective harvesting. These arguments support the "Complete date" and the "Complete date plus hours, minutes and seconds" granularities defined in ISO The legitimate formats are and Both arguments must have the same granularity. All repositories must support . A repository that supports should indicate so in the response. A request by a harvester with finer granularity than that supported by a repository must produce an error.

UTCdatetime in Protocol Responses

Datestamps appear in the headers of records that are returned in response to , and requests. These datestamps are encoded using ISO and are expressed in UTC; they must be expressed in the finest granularity supported by the repository. The value of the datestamp must correspond to the rules for datestamp-based selective harvesting.

Each protocol response includes a element, which must be the time and date of the response in UTC. This is encoded using the "Complete date plus hours, minutes, and seconds" variant of ISO This format is .

A in a protocol reply may include an optional argument , which is expressed in UTC. This is encoded using the "Complete date plus hours, minutes, and seconds" variant of ISO This format is .

metadataPrefix and Metadata Schema

OAI-PMH supports the dissemination of records in multiple metadata formats from a repository. The request returns the list of all metadata formats available from a repository, each of which has the following properties:

• The - a string to specify the metadata format in OAI-PMH requests issued to the repository. consists of any valid URI unreserved characters. arguments are used in , , and requests to retrieve records, or the headers of records that include metadata in the format specified by the ;
• The metadata schema URL - the URL of an XML schema to test validity of metadata expressed according to the format;
• The XML namespace URI that is a global identifier of the metadata format.

The metadata in each record returned by and must comply with the conventions of the XML namespace specification. This means that the root element of the metadata part must contain an attribute, the value of which is the XML namespace URI of the metadata format. The root element must also contain an attribute that has a value that includes the URL of the XML schema for validation of the metadata. This URL must match the URL of the metadata schema for the included as an argument to the or request (the mapping from to metadata schema is defined by the repository's response to the request).

For purposes of interoperability, repositories must disseminate Dublin Core, without any qualification. Therefore, the protocol reserves the `', and the URL of a metadata schema for unqualified Dublin Core, which is manicapital.com The corresponding XML namespace URI is manicapital.com

The `' is reserved for future use. Implementations should not use this .

Communities should adopt guidelines for sharing of metadata schema and XML namespace URI's of metadata formats. Such guidelines are outside of the scope of the OAI-PMH. The accompanying Implementation Guidelines document provides some sample XML Schema and instance documents for common metadata formats such as MARC and RFC

Flow Control

A number of OAI-PMH requests return a list of discrete entities: returns a list of records, returns a list of headers, and returns a list of sets. Collectively these requests are called list requests. In some cases, these lists may be large and it may be practical to partition them among a series of requests and responses. This partitioning is accomplished as follows:

• A repository replies to a request with an incomplete list and a
• In order to make the response a complete list, the harvester will need to issue one or more requests with as arguments. The complete list then consists of the concatenation of the incomplete lists from the sequence of requests, known as a list request sequence.

Details of flow control and the are as follows:

• The only defined use of is as follows:
  • a repository must include a element as part of each response that includes an incomplete list;
  • in order to retrieve the next portion of the complete list, the next request must use the value of that element as the value of the argument of the request;
  • the response containing the incomplete list that completes the list must include an empty element;
  All other uses of by a harvester are illegal and must return an error.
• In all cases when a is issued, the incomplete list must consist of complete entities; e.g., all individual records returned in an incomplete record list from a request must be intact.
• The format of the is not defined by the OAI-PMH and should be considered opaque by the harvester.
• The protocol does not define the semantics of incompleteness. Therefore, a harvester should not assume that the members in an incomplete list conform to some selection criteria (e.g., date ordering).
• Before including a in the URL of a subsequent request, a harvester must encode any special characters in it.

The following optional attributes may be included as part of the element along with the itself:

• -- a UTCdatetime indicating when the ceases to be valid.
• -- an integer indicating the cardinality of the complete list (i.e., the sum of the cardinalities of the incomplete lists). Because there may be changes in a repository during a list request sequence, as described under Idempotency of resumptionTokens, the value of may be only an estimate of the actual cardinality of the complete list and may be revised during the list request sequence.
• -- a count of the number of elements of the complete list thus far returned (i.e. starts at 0).

The following example is a series of ListRecords requests where the complete list consists of records and the repository only returns records per response.

• The harvester issues a request.
• The repository responds with an incomplete list of records. The repository marks this list as incomplete by including in the response a non-emptyelement, with two attributes: a of , and a of 0.
• The harvester issues a subsequent request that includes the that it received in the previous response.
• The repository responds with an incomplete list of 75 records. The repository marks this list as the final incomplete list by including in the response an empty element with two attributes: a of , and a of

This flow control mechanism, in combination with HTTP transport layer facilities, provides some basic tools with which a repository can enforce an acceptable use policy for its harvesting interface. Communities implementing the OAI-PMH may need more extensive tools to enforce acceptable use policies for either the harvesting interface of their repositories or for the metadata harvested from those repositories. The enforcement of such additional policies is outside of the scope of the OAI-PMH.

Idempotency of resumptionTokens

Repositories that implement must do so in a manner that allows harvesters to resume a sequence of requests for incomplete lists by re-issuing a list request with the most recent . The purpose of this is to allow harvesters to recover from network or other errors that would otherwise mean that the list request sequence would have to be started again. A re-issue of a list request with a occurs in two contexts:

1. When there are no changes in the repository. There are no changes to the complete list returned by the list request sequence. In this case, the repository must return the same incomplete list when the most recent list request, i.e. the one with the most recent non-expired , is re-issued.
2. When there are changes in the repository. There may be changes to the complete list returned by the list request sequence. These changes occur when the records disseminated in the list move in or out of the datestamp range of the request because of changes, modifications, or deletions in the repository. In this case, strict idempotency of the incomplete-list requests using values is not required. Instead, the incomplete list returned in response to a re-issued request must include all records with unchanged within the range of the initial list request. The incomplete list returned in response to a re-issued request may contain records with datestamps that either moved into or out of the range of the initial request. In cases where there are substantial changes to the repository, it may be appropriate for a repository to return a error, signaling that the harvester should restart the list request sequence.

Error and Exception Conditions

In event of an error or exception condition, repositories must indicate OAI-PMH errors, distinguished from HTTP , by including one or more elements in the response. While one element is sufficient to indicate the presence of the error or exception condition, repositories should report all errors or exceptions that arise from processing the request. Each element must have a attribute that must be from the following table; each element may also have a free text string value to provide information about the error that is useful to a human reader. These strings are not defined by the OAI-PMH.

Error Codes	Description	Applicable Verbs
The request includes illegal arguments, is missing required arguments, includes a repeated argument, or values for arguments have an illegal syntax.	all verbs
The value of the argument is invalid or expired.
Value of the argument is not a legal OAI-PMH verb, the verb argument is missing, or the argument is repeated.	N/A
The metadata format identified by the value given for the argument is not supported by the item or by the repository.
The value of the argument is unknown or illegal in this repository.
The combination of the values of the , , and arguments results in an empty list.
There are no metadata formats available for the specified item.
The repository does not support sets.

The following example demonstrates error handling in the case of an illegal verb argument. All request URLs shown from now on will be wrapped to make them more readable.

Request

manicapital.com? verb=nastyVerb

Response

<?xml version="" encoding="UTF-8"?> <OAI-PMH xmlns="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <responseDate>TZ</responseDate> <request>manicapital.com</request> <error code="badVerb">Illegal OAI verb</error> </OAI-PMH>

The following example demonstrates error handling in the case of a request to a repository that does not handle sets.

Request

manicapital.com? verb=ListSets

Response

<?xml version="" encoding="UTF-8"?> <OAI-PMH xmlns="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <responseDate>TZ</responseDate> <request verb="ListSets">manicapital.com</request> <error code="noSetHierarchy">This repository does not support sets</error> </OAI-PMH>

This section lists the requests, or defined in the OAI-PMH. The documentation for each request is organized as follows:

• A section title corresponding to the token used to specify the request as the required argument to an HTTP request.
• A brief summary of the meaning of the verb and notes on its usage.
• The list of additional arguments for the request. Arguments are of three types:
  • required, the argument must be included with the request (the argument is always required, as described in HTTP Request Format).
  • optional, the argument may be included with the request.
  • exclusive, the argument may be included with request, but must be the only argument (in addition to the argument).
• Error and exception conditions specific to the protocol request.
• One or more example requests and corresponding responses, with explanatory notes if appropriate.

An XML Schema defines the format of valid replies to all OAI-PMH requests.

GetRecord

Summary and Usage Notes

This verb is used to retrieve an individual metadata record from a repository. Required arguments specify the identifier of the item from which the record is requested and the format of the metadata that should be included in the record. Depending on the level at which a repository tracks deletions, a header with a "deleted" value for the attribute may be returned, in case the metadata format specified by the is no longer available from the repository or from the specified item.

Arguments

• a required argument that specifies the unique identifier of the item in the repository from which the record must be disseminated.
• a required argument that specifies the of the format that should be included in the metadata part of the returned record . A record should only be returned if the format specified by the can be disseminated from the item identified by the value of the identifier argument. The metadata formats supported by a repository and for a particular record can be retrieved using the request.

Error and Exception Conditions

• - The request includes illegal arguments or is missing required arguments.
• - The value of the argument is not supported by the item identified by the value of the argument.
• - The value of the argument is unknown or illegal in this repository.

Examples

Request

Request a record in the Dublin Core metadata format [URL shown without encoding to be more readable].

manicapital.com? verb=GetRecord&identifier=oai:manicapital.com:cs/&metadataPrefix=oai_dc

Response

<?xml version="" encoding="UTF-8"?> <OAI-PMH xmlns="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <responseDate>TZ</responseDate> <request verb="GetRecord" identifier="oai:manicapital.com:cs/" metadataPrefix="oai_dc">manicapital.com</request> <GetRecord> <record> <header> <identifier>oai:manicapital.com:cs/</identifier> <datestamp></datestamp> <setSpec>cs</setSpec> <setSpec>math</setSpec> </header> <metadata> <oai_dc:dc xmlns:oai_dc="manicapital.com" xmlns:dc="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <dc:title>Using Structural Metadata to Localize Experience of Digital Content</dc:title> <dc:creator>Dushay, Naomi</dc:creator> <dc:subject>Digital Libraries</dc:subject> <dc:description>With the increasing technical sophistication of both information consumers and providers, there is increasing demand for more meaningful experiences of digital information. We present a framework that separates digital object experience, or rendering, from digital object storage and manipulation, so the rendering can be tailored to particular communities of users. </dc:description> <dc:description>Comment: 23 pages including 2 appendices, 8 figures</dc:description> <dc:date></dc:date> </oai_dc:dc> </metadata> </record> </GetRecord> </OAI-PMH>

Request

Request a record in the Dublin Core metadata format. The requested record, however, can not be returned because the identifier does not exist. Therefore, the response does not contain a container. It does have an element with a attribute that has the value . [URL shown without encoding for better readability].

manicapital.com? verb=GetRecord&identifier=oai:manicapital.com:quant-ph/&metadataPrefix=oai_dc

Response

<?xml version="" encoding="UTF-8"?> <OAI-PMH xmlns="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <responseDate>TZ</responseDate> <request verb="GetRecord" identifier="oai:manicapital.com:quant-ph/" metadataPrefix="oai_dc">manicapital.com</request> <error code="idDoesNotExist">No matching identifier in arXiv</error> </OAI-PMH>

Request

Request a record in the oai_marc metadata format. However, the requested metadata format can not be disseminated for this identifier. Therefore, the response contains no record. It does contain an element with a attribute that has the value . [URL shown without encoding for better readability].

manicapital.com? verb=GetRecord&identifier=oai:manicapital.com:quant-ph/&metadataPrefix=oai_marc

Response

<?xml version="" encoding="UTF-8"?> <OAI-PMH xmlns="manicapital.com" xmlns:xsi="manicapital.com" xsi:schemaLocation="manicapital.com manicapital.com"> <responseDate>TZ</responseDate> <request verb="GetRecord" identifier="oai:manicapital.com:quant-ph/" metadataPrefix="oai_marc">manicapital.com</request> <error code="cannotDisseminateFormat"/> </OAI-PMH>

Identify

Summary and Usage Notes

This verb is used to retrieve information about a repository. Some of the information returned is required as part of the OAI-PMH. Repositories may also employ the Identify verb to return additional descriptive information.

Arguments

None

Error and Exception Conditions

• - The request includes illegal arguments.

Response Format

The response must include one instance of the following elements:

• : a human readable name for the repository;
• : the base URL of the repository;
• : the version of the OAI-PMH supported by the repository;
• : a UTCdatetime that is the guaranteed lower limit of all datestamps recording changes, modifications, or deletions in the repository. A repository must not use datestamps lower than the one specified by the content of the element. must be expressed at the finest granularity supported by the repository.
• : the manner in which the repository supports the notion of deleted records. Legitimate values are ; ; with meanings defined in the section on deletion.
• the finest harvesting granularity supported by the repository. The legitimate values are and with meanings as defined in ISO

The response must include one or more instances of the following element:

• : the e-mail address of an administrator of the repository.

The response may include multiple instances of the following optional elements:

• : a compression encoding supported by the repository. The recommended values are those defined for the header in Section of RFC describing HTTP A element should not be included for the encoding, which is implied.
• : an extensible mechanism for communities to describe their repositories. For example, the container could be used to include collection-level metadata in the response to the Identify request. Implementation Guidelines are available to give directions with this respect. Each container must be accompanied by the URL of an XML schema describing the structure of the description container.

Examples

Request

manicapital.com? verb=Identify

Response

The below example of a response to the request contains three containers:

• The container complies to an XML Schema, which is available at manicapital.com This schema, provided in the accompanying Implementation Guidelines document, is used by repositories that choose to comply with a specific format of unique identifiers for items. The format of that identifier is explained by means of comments in the manicapital.com XML Schema.
• The container complies to an XML Schema, which is available at manicapital.com This schema, provided in the accompanying Implementation Guidelines document, has been agreed upon by the OAI e-print community, and contains information specific to repositories in that community.
• The container complies to an XML Schema, which is available at manicapital.com This schema, provided in the accompanying Implementation Guidelines document, is used by repositories that want to point harvesters to other repositories, by listing their base URLs. Usage of the container is recommended; it may support harvesters in discovering the network-location of repositories.